xanaduai / qhack2022 Goto Github PK
View Code? Open in Web Editor NEWQHack—The one-of-a-kind quantum computing hackathon
Home Page: https://qhack.ai
QHack—The one-of-a-kind quantum computing hackathon
Home Page: https://qhack.ai
Lindwurm
Quantum sea - Classifying water molecules and sodium ions in protein structures
The goal of the project is building quantum machine learning-based classifiers which can classifies water molecules and sodium ions present in the crystallographic structure of protein obtained by X-ray crystallography, as a kind of toy program for predicting physicochemical properties related with the protein. X-ray crystallography is mainly used to obtain the structure of a protein with high resolution, by using diffraction of X-ray due to electrons in the protein. Due to the nature of the method, small molecules, atoms or ions with the same number of electrons are likely to produce similar peaks. For example, water molecule, one of the small molecules abaundant in protein crystal structures, have 10 electrons, is likely to be confused with sodium ions which has 10 electrons. However, since water molecules does not have net charge, while sodium ions having positive net charge, the structure of proteins that can hold water molecules and sodium ions are likely to be different. From this, water molecules and sodium ions in X-ray crystallography can be distinguished.
In this project, convolutional natural network-based water-sodium ion classifier with input as a voxelized 3D image of the structure of carbon, nitrogen, and oxygen atoms from proteins or other compounds (exclude water) in a cube which center is located at a location where sodium ion or water molecule exists and size of 16Å and grid spacing of 0.5Å. In the last layer before fully connected layer of the classifier, trainable quanvolution neural network was used for
convolution and pooling of 2x2x2 grid a into 1x1x1 grid. With the help of IBM's 16-qubit quantum computer, the performance between 2x2x2 quanvolutional layer implemented as 2 circuits of 4-qubit quanvolution circuit and one circuit of 8-qubit quanvolution can be compared. Furthermore, 16-qubit circuit quanvolutional neural network as composed layer consists with convolution of dilated convolution and non-dilated convolution can be tested.
https://github.com/shadow1229/Qhack_2022/tree/main/Quantum_sea
Training will be done by quantum simulator, and IBM's Quantum Computer is expected to be used in the comparing the performance between classifiers. From the currently implemented method, it is expected that a 4-qubit circuit with four random rotations and two layers will be used 13,824,000 shots. (108 input structures * 64 channels * 2 circuits * 1000 shots per exposure value) In similar mannner, evaluation of 8-qubit circuit and 16-qubit circuit are expected to use 6,912,000 shots each.
Your team's name (matching the name used on the QHack Coding Challenges, if applicable)
A brief description of your project (1-2 paragraphs).
As in the public news, the first Quantum Satellite was successfully launched in 2016, and the first Quantum Communication was carried out in 2017 between Austria and China. Recently the entanglement between 2 places with a distance over 1000 kilometers was detected. This project is to attempt integration Quantum Computers with Quantum Communication.
My high school teacher was actually highly encouraging and recommending me to study theoretic physics with my excellent school performance and strong interest in physics,. After my honorable graduation from Tongji Medical School at Huazhong University of Science and Technology and my Residency on Internal Medicine at Union (Xie He) Hospital. I went to Germany and did my doctoral thesis at Munich University in the interest of science. After obtaining my doctoral degree with honor from Munich University, I came to University of California at Los Angeles (UCLA) for my post-doctoral fellowship and subsequently worked there. I also took and successfully passed the U.S. National Board Step-1, National Board Step-2 and Clinical Skill Assessment (CSA), and am certified by the U.S. Educational Commission for Foreign Medical Graduates (ECFMG).
Part of my post-doctoral fellowship at UCLA was working with an engineer who designed an electronic device controlled automatically by computers to work with biologic systems, for which I went back to a community college in Los Angeles on my own to learn Fortran. In 2006, I moved from Los Angeles (UCLA) to Washington, DC for my job at the National Institutes at Health (NIH). I have strengthened my interests in physic and also expanded my areas in technologies, including quantum communication and quantum computers. My e-mail is [email protected] and my phone # is 240-453-1534. I am looking forward to hearing from you about my draft of this QHack Open Hackathon project.
Many thanks, Sincerely,
Yining Xie
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc).
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
https://doi.org/10.1145/3498691
Two Attachments here: 3498691.pdf; appendices.pdf
3498691.pdf
appendices.pdf
If awarded, the access to IBM Quantum machine with IBM 16-qubit QPU will be used to finish this project.
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc) as stated the above.
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
Your team's name (matching the name used on the QHack Coding Challenges, if applicable)
A brief description of your project (1-2 paragraphs).
Severe acute respiratory syndrome coronavirus 2 (SARS CoV 2) for COVID-19, including its variants, has been wide spread globally. It affects all ages, races and various medical conditions. Using traditional computers is quite challenging to analyze interactions between proteome and genome, including gene-gene Interactions among virus and human genetics. This project is to use currently available quantum computers to analyze interactions between proteome and genome, including gene-gene interactions in SARS CoV 2 and human genetics.
Virology and genetics have been two fields that I’m very interested since my medical school. After my honorable graduation from Tongji Medical School at Huazhong University of Science and Technology where I learned Western Medicine and some Chinese Medicine, I did my Residency on Internal Medicine at Union (Xie He) Hospital. I then went to Germany and did my doctoral thesis at Munich University. After obtaining my doctoral degree with honor from Munich University, I came to University of California at Los Angeles (UCLA) for my post-doctoral fellowship and subsequently worked there. I also took and successfully passed the U.S. National Board Step-1, National Board Step-2 and Clinical Skill Assessment (CSA), and am certified by the U.S. Educational Commission for Foreign Medical Graduates (ECFMG).
In 2006, I moved from Los Angeles (UCLA) to Washington, DC for my job at the National Institutes at Health (NIH), and further strength my interests in Virology and genetics, particular at the time of current COVID-19 pandemics. My e-mail is [email protected] and my phone # is 240-453-1534. I am looking forward to hearing from you about my draft of this QHack Open Hackathon project.
Many thanks, Sincerely,
Yining Xie
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc)
https://doi.org/10.1145/3498691
Two Attachments here: 3498691.pdf; appendices.pdf
3498691.pdf
appendices.pdf
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
If awarded, the access to IBM Quantum machine with IBM 16-qubit QPU will be used to finish this project.
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc).
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the additional AWS credits, if awarded, to finish your Open Hackathon project.
QRiders
Solanaceous crops(Potato, Tomato, and Pepper) are one of the most cultivated and in-demand crops after rice and wheat. Solanaceous farming dominates as an occupation in the agriculture domain in more than 125 countries. However, even these crops are, subjected to infections and diseases, mostly categorized into two grades:
Farmers who grow these crops suffer from serious financial standpoint losses each year. Wouldn't it be amazing if the farmer could discover this ailment early and treat it properly? This can prevent a lot of waste and financial loss.
https://github.com/bopardikarsoham/QHack-Open-Hackathon
If I get access to the 16 qubit IBM machine, I'll be able to encode somewhat larger images than a standard 2x2 or 4x4. It has the potential to significantly improve my project and understanding of QML algorithms on larger qubit devices.
talll
We demonstrate a variational quantum autoregressive model (VQAR) to forecast stochastic data trends of foreign exchange rates and COVID-19 cases.
Readme
Code repository is found at repo
Higher AR memory and joint data prediction across countries/states require more qubits.
16-qubit QPU would be a good size for forex/covid prediction.
Our project is qualified for
MQS
In a VQE algorithm, the parameters of the variational circuit are updated by the classical optimizer. Classical optimizers rely on calculating the gradient of the cost function at each iteration. In the case of applying VQE to find the ground state of a molecule, the cost function is the expectation value of the Hamiltonian of the molecular system. There are multiple ways to calculate the gradient of a function, and one of the best ways to do this is using the parameter shift rule. In the most commonly studied case of finding the ground state energy of a molecule, the Hamiltonian itself is independent of the parameters θ. For some models, such as PCM-VQE, which
models the ground state energy of a molecule surrounded by an implicit solvent, this is no longer true. In such a model, the Hamiltonian itself is modified by the electronic wavefunction because the electronic structure interacts with the solvent, meaning that we have a parameter dependent Hamiltonian. This means that the parameter-shift rule needs to be modified in order to be used.
Therefore, for situations with a parameter-dependent Hamiltonian, a custom parameter-shift rule needs to be derived. In the code provided in the PCM-VQE preprint, the finite difference method is used for gradient calculation. In the paper it is mentioned that it has not been yet determined how the parameter-dependence of the Hamiltonian affects the convergence properties of the VQE algorithm. As a first step to exploring this topic, it would be beneficial to replace the finite-difference method with a parameter-shift rule. This would disentangle the approximation errors from the genuinely novel effects that the simulation of this model carries with it. To finally assess the convergence properties, there are lots of variables to explore, such as initial parameter values, size of orbital basis set and the gate composition of the circuit.
pdf version with equations and more information: https://github.com/MQSdk/parameter_shift_H_theta/blob/main/description.pdf
https://github.com/MQSdk/parameter_shift_H_theta
We will use the AWS credits to evaluate the convergence properties on different quantum computers (IONQ and Rigetti) and do a thorough analysis of different circuit implementations and initial parameter settings. Further, we will also increase the number of orbitals in the basis set which requires increasingly more qubits to evaluate the threshold of possible chemical systems to run on the quantum hardware available in AWS Braket.
Members: @StreakSharn, @DSamuel1, @r-agni.
Estimating how to price Financial Derivatives - like options such as puts and calls - is a difficult task due to the huge number of possible changes in variables. While estimation techniques such as Classical Monte Carlo exist, they can easily rack up large 'error' or uncertainty; getting rid of this is time-consuming and costly.
In the case of the Classical Monte Carlo, for example, error scales with 1/sqrt(M) where M is the number of simulations. Because of this, in order to halve the error, you must quadruple the simulation number. To reduce the error to useful amounts, the quadratic scaling can mean large numbers of simulations are needed.
In Quantum Monte Carlo, however, we can offer a Quadratic Speedup, so error scales with 1/M. This has huge potential, since it can greatly improve the accuracy of Option Pricing while reducing the intensity of simulation required for them.
Team Quest hopes to explore this by implementing work in Quantum computational finance: Monte Carlo pricing of financial derivatives, seeing how Quantum Monte Carlo can be realised and executed.
At small scales with few qubits, the Quantum Monte Carlo offers no tangible speed-up or improvement in error over its Classical counterpart - in fact, due to the inherent errors present within NISQ systems, even with error correction it often comes out worse. However, at larger scales, the advantage gained from Quantum methods can lead it to outstrip the Classical equivalent.
IBM's 16-qubit QPU would allow us to approach these scales, and more effectively demonstrate advantages offered by Quantum Monte Carlo algorithms.
This could allow us to expand the breadth of our Project, and examine/evaluate the comparison between Quantum and Classical methods.
QuantumRing
Neutrino Oscillation which was first predicted by Bruno Pontecorvo in 1957, has been a great theoretical and experimental interest, as its precise properties can shed light on several properties of the neutrino. The experimental discovery of neutrino oscillation proves that neutrino has non-zero mass, which then causes a required modification to the Standard Model of particle physics. This experimental work by Takaaki Kajita and Arthur B. McDonald was so great that it was recognized with the 2015 Nobel Prize for Physics. Due to the potential of this process, many attempts have been conducted to gain a profound understanding of the phenomenon. In this project, we try to dive in and explore this complex quantum dynamic but with a different approach.
Using a quantum computer, our team try to simulate collective flavor oscillations which are created by the interaction neutrino-neutrino in a neutrino cloud with a high density of neutrinos. This process can happen in supernovae and the early universe - astrophysical scenarios with large neutrino density. In this project, we are considering a two-flavor case of interacting neutrinos which leads us to demonstrate the time and space evolution of the set of amplitudes from a Schrodinger equation:
The H - Hamiltonian from the equation is the Hamiltonian for neutrino flavor evolution in an environment with a high density of neutrinos which include vacuum and forward-scattering interaction contributions. Here we use QAOA (Quantum Approximate Optimization Algorithm) to realize this Hamiltonian with the ambition to scale the system to as many neutrinos as possible, this is where we need a quantum computer with many qubits to perform our quantum circuit. From this simulation, we aim to find a way to simulate many neutrinos interacting systems with polynomial scale-up which will become a great tool for researchers and scientists to look into this complex neutrino dynamic.
https://github.com/bachbao/Simulating-collective-neutrino-oscillation-using-QAOA-algorithm
In this project, we plan to use the power of a real quantum computer, therefore a great need for resources from a real quantum computer is what we want. By implementing the QAOA to realize the neutrino Hamiltonian, we will need to use quantum tomography which means running multiple circuits with many shots to retrieve the initial state. That's our first reason, we need resources from AWS to gain the ability to run many jobs on the Rigetti and Ionq quantum computers, to be more specific Ionq quantum computing can help us a lot with its fully connected machine as we will use many Toffoli gates across our circuits. The second reason is in this current generation of quantum computers, noisy, near-term, intermediate size, it is important to check whether our method can work well using a real quantum computer as the project will lose most of its application meaning if it is only presented by using the simulator. This comes to our final and most important reason for demanding the resources from AWS power up, we plan to scale up our system to as many neutrinos as possible, therefore we want to have a large-scale quantum computer and the quantum computers from Rigetti and Ionq satisfy our need.
Benjamin Hall, Alessandro Roggero, Alessandro Baroni, Joseph Carlson. "Simulation of Collective Neutrino Oscillations on a Quantum Computer." arXiv preprint arXiv:2102.12556 (2021).
Zewei Xiong. "Many-body effects of collective neutrino oscillations" arXiv preprint arXiv:2111.00437 (2021).
How do proteins fold? One of the million-dollar questions that we haven’t known the answer for. Understanding why a particular protein only folds that way and why not the other hundreds of thousands of possibilities can take homo sapiens a long way. To invent new drugs and medicines, we need the protein’s structure and that is not that simple and easy. A moderate amino acid chain protein itself can have a conformation size of the mass of the Earth. With the increase in the amino acids in the chain, the potential conformations grow exponentially which is why this problem is hard to tackle using a classical computer.
Alzheimer's disease is a dreadful disease caused in humans where it destroys memory and other important mental functions. In India, over 1 million people suffer from Alzheimer’s disease every year. The main objective of this project is to use quantum computing and find the ground state of the BACE1 exosite binding peptide because almost all of these peptides fold into their ground states. We would love to use different techniques such as Error Mitigation to compare and check the results.
The resource estimate to complete this project will be around $1000. With this resource, we will be able to access different quantum computers by various companies, so that we can run the experiments on real hardware and bring out the best result finally. We would also like to check the results with different optimizers, algorithms and compare the results showcasing the algorithm and optimizer that performed very well for this particular problem.
Amazon Bracket Challenge
IBM Qiskit Challenge
Bio-QML Challenge
Quantum Chemistry Challenge
The Superpositioned States of America
Our work focuses on Quantum Graph Neural Networks (QGNNs), to solve the particle tracking reconstruction challenge. Specifically, we are looking to focus on the detailed analysis of the vanishing gradient problem, long training times, and how robust the overall approach is to noise from real quantum computers, which have been mentioned but not addressed yet in prior work. Our work aims to improve the viability of the QGNN method for particle tracking problems.
Optimizing Quantum Graph Neural Networks for the Particle Tracking Problem.
Currently, our team has secured initial power-ups in the form of Amazon Braket credits and access to a 7-qubit IBM quantum machine. Original QEN circuit used in the A Quantum Graph Neural Network Approach to Particle Track Reconstruction utilizes 7 qubits. However, the QNoN circuit uses 11 qubits, which is why having access to IBM’s 16-qubit machine would be helpful for benchmarking purposes.
SexyQuantumGuys
Classical Neural Networks (NNs) have been used in various ways to predict the ground state of molecules [1]. In such NN-based schemes, atom-centered symmetry functions (usually referred to as ‘descriptors’) are employed to ensure the translational, rotational, and permutation invariance of the many-body system. Considering the quantum version of this approach, we aim to achieve higher accuracy in predicting the ground state of molecules in shorter time, using Quantum Hybrid Neural Networks based on novel quantum descriptors with learnable Hamiltonian.
Also if possible, we are trying to make a Hamiltonian for the Bond Order Potential(BOP). Bonding potential is the largest part of molecule's energy. So by making a nice bonding Hamiltonian, the energy prediction will be much better. The energy will be predicted with the QAOA Method.
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
https://github.com/justids/SexyQuantumGuys
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
To make more accurate prediction, we need to use as many as possible qubits for the descriptors. So we will try up to 12 qubits to predict it well.
###Challenges
Amazon Braket Challenge
IBM Qiskit Challenge
Hybrid Algorithms Challenge
QAOA Challenge
Quantum Chemistry Challenge
Simulation Challenge
Young Scientist Challenge
###Reference
Jörg Behler
Chemical Reviews 2021 121 (16), 10037-10072
DOI: 10.1021/acs.chemrev.0c00868
Samras
Quantum entanglement used as a resource adds an advantage that cannot be obtained with purely classical approaches. This phenomenon manifests itself in CHSH game and quantum superdense coding, where entanglement boosts the winning probabilities for the players and communication rate, respectively.
In our project, we investigate the use of entanglement in the task of quantum autoencoding. Quantum autoencoders [1] compress quantum information into smaller dimensional Hilbert spaces, and moreover, it is known that entanglement resources can aid in this compression [2]. We take state-of-the-art quantum autoencoder strategies, and add additional entanglement resources to test for a compression advantage. We aim to test the approach on datasets such that the known autoencoder schemes do not compress and decompress the data properly, but when we add the entanglement resources, the result improves. Indeed part of our project is to identify such datasets - at this point we are working with the breast cancer data set and with the credit card fraud dataset.
We believe that use-cases beyond current research directions might arise with regards to entanglement-assisted quantum autoencoders. They might play a role in quantum state transmission, where one party physically transmits a quantum state to another party. The two parties, as in superdense coding and the CHSH game, share entanglement ahead of time. The sending party uses her share of entanglement to aid in compressing the total state she wants to transmit and then transmits. The receiver uses his share to decompress the state and potentially de-noise his received quantum states. Such scenarios already arise in multi-processor quantum computing, where quantum states move between various quantum processors, or in schemes like quantum key distribution, where some schemes like E91 and DI-QKD, already deploy quantum entanglement as a resource.
[1] Romero, Jonathan, Jonathan P. Olson, and Alan Aspuru-Guzik. "Quantum autoencoders for efficient compression of quantum data." Quantum Science and Technology 2.4 (2017): 045001.
[2] Z. B. Khanian and A. Winter, "Entanglement-Assisted Quantum Data Compression," 2019 IEEE International Symposium on Information Theory (ISIT), 2019, pp. 1147-1151, doi: 10.1109/ISIT.2019.8849352.
https://github.com/VoicuTomut/temporaryRepoQhack
We intend to run experiments on QPUs for the validation or falsification of our concept. We will start with 4 qubits encoding MNIST dataset images using either angle (2 by 2 pixelated pictures) or amplitude encoding (4 by 4 pixelated pictures). The amount of entanglement qubits needed to gain an advantage is not yet known, as we are currently working towards preliminary results showing advantage. AWS Braket offers simulation platforms like SV1 which are high-performance and can simulate up to 34 qubits which we believe will provide just enough qubit resources to test our approach (4 data qubits, reference qubits, a qubit for swap test, entanglement resources qubits, additional auxiliary qubits). Later on we would like to work with breast cancer and finance datasets. For each we would need around 1000 backends runs with 2500 each. Our simulation results demonstrate that quantum auto encoders might serve as a great tool for anomaly detection for the above-mentioned cases.
SexyQuantumGuys
Classical Neural Networks (NNs) have been used in various ways to predict the ground state of molecules [1]. In such NN-based schemes, atom-centered symmetry functions (usually referred to as ‘descriptors’) are employed to ensure the translational, rotational, and permutation invariance of the many-body system. Considering the quantum version of this approach, we aim to achieve higher accuracy in predicting the ground state of molecules in shorter time, using Quantum Hybrid Neural Networks based on novel quantum descriptors with learnable Hamiltonian.
Also if possible, we are trying to make a Hamiltonian for the Bond Order Potential(BOP). Bonding potential is the largest part of molecule's energy. So by making a nice bonding Hamiltonian, the energy prediction will be much better. The energy will be predicted with the QAOA Method.
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
https://github.com/justids/SexyQuantumGuys
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
To make more accurate prediction, we need to use as many as possible qubits for the descriptors. So we will try up to 12 qubits to predict it well.
###Challenges
Amazon Braket Challenge
IBM Qiskit Challenge
Hybrid Algorithms Challenge
QAOA Challenge
Quantum Chemistry Challenge
Simulation Challenge
Young Scientist Challenge
###Reference
Jörg Behler
Chemical Reviews 2021 121 (16), 10037-10072
DOI: 10.1021/acs.chemrev.0c00868
QSchrodinger
genetic quantum algorithm (GQA). GQA is based on the concept and principles of quantum computing such as qubits and superposition of states. Instead of binary, numeric, or symbolic representation, by adopting qubit chromosome as a representation GQA can represent a linear superposition of solutions due to its probabilistic representation. As genetic operators, quantum gates are employed for the search of the best solution. Rapid convergence and good global search capability characterize the performance of GQA. The effectiveness and the applicability of GQA are demonstrated by experimental results on the knapsack problem, The results show that GQA is superior to other genetic algorithms using penalty functions, repair methods, and decoders.
a quantum genetic algorithm consist on:
1- First step, the algorithm begins preparing a superposition of all individuals, i.e., N, or chromosomes of population Q(t), Therefore, all individuals are represented by only one individual quantum register. That is, the entire population is represented by a single chromosome in a superposition state, One of the key steps of RQGA is the correlation between the individual quantum register |x>i and a fitness quantum register | f itness>i.
2- In a second step the algorithm searches for the maximum fitness. Once the operator F is applied, RQGA searches for the maximum fitness value based on the Grover’s search algorithm.
3- Finally, making a measure in the chromosome with maximum fitness is obtained.
Reference material:
https://www.mdpi.com/2073-431X/5/4/24/pdf
https://arxiv.org/pdf/cs/0403003.pdf
https://www.researchgate.net/publication/2573070_Genetic_Quantum_Algorithm_and_its_Application_to_Combinatorial_Optimization_Problem
https://github.com/echchallaouy/Quantum-genetics
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
We are going to use the IBM 16-qubit QPU, if awarded to simulate a large number of the population's chromosomes.
1- In the first place, Genes are the qubits.
2- Second, preparing the superposition of all individuals, or chromosomes of the population.
3- In third place, the oracle O marks the maximum fitness of |ψ>i, such that when the oracle is applied we obtain the superposition, this step is repeated a given number of iterations. The Grover’s maximum number of iterations is calculated as pi *sqrt(2**n)/4 where n is the number of qubits or length of the quantum chromosome.
4- In fourth and last place the Grover’s diffusion operator G finds the chromosome with a marked state.
5- Finally, making a measure, we get the state that points to chromosome with maximum fitness
UncertaintyHack
We implement a quantum-classical hybrid QLSTM model by incorporating quantum variational layers into the classical LSTM in order to improve the efficiency and trainability of LSTM for better stock price prediction.
Stock price prediction is one of the most rewarding problems in modern finance, where the accurate forecasting of future stock prices can yield significant profit and reduce the risks. LSTM (Long Short-Term Memory) is a recurrent Neural Network (RNN) applicable to a broad range of problems aiming to analyze or classify sequential data. LSTM can be used to predict the future stock price based on the historical data sequences. Recent studies have shown that its efficiency and trainability can be improved by leveraging a quantum-classical hybrid model of LSTM. QLSTM proved to learn significantly more information after the first training epoch than its classical counterpart. Thus, we implement a variational quantum-classical hybrid algorithm engaging machine learning techniques within the LSTM model framework in order to predict stock price movement.
Firstly, we conduct the feature preparation of the stock price using data collection of the technical indicators for a given stock, correlated assets, the sentiment analysis of the related sources of the information from the market: news, social media, reports. Then we test a classical multi-input LSTM on the prepared dataset and analyze the feasibility of this instrument to predict the prices. Finally, we modify classical LSTM by introducing variational quantum circuits and comparing the result to that of classical LSTM.
###Viability of the algorithm
QLSTM shows better prediction accuracy (10 times less RMSE) and trainability compared to classical LSTM as shown by Fang et. al (2020) Furthermore, QLSTM requires a relatively low amount of qubits. The depth of the variational quantum circuit (d) grows linearly both on the number of variational layers and the number of qubits. An iterative optimization technique is used to update the variational parameters of the quantum circuit employed in the hidden layer of the Neural Network in a way, that each of the learned parameters can effectively absorb the surrounding noise without even knowing any properties of the noise. Low gate depth per run and lower qubit count, together with noise tolerance make the variational technique viable to use in NISQ. (Noisy Intermediate-Scale) era devices.
Please refer to our GitHub repo here
For better trainability of LSTM, we need to try different numbers of variational layers, qubits, and circuit depth. As such, we will run more trials and use IBM quantum computers with a greater number of qubits.
Parmanu
The goal of the project is to explore the Sub Space Search VQE Algorithm to calculate the Kth excited-state energy of a given hamiltonian. There are three variants of SSVQE shown below:
This project aim towards comparing the results of all three algorithms and how they perform on noiseless and noisy systems.
The current tests have been done using the Hamiltonian of the H2 molecule. which alone requires 4 qubits. We want to try these algorithms for relatively large or complex molecules to obtain an idea about how they perform on noisy channels and possibly try to apply Error Mitigation techniques to balance it out. For that, we would like to have access to the 16-Qubit system.
To implement an algorithm for determining the vibrionic structure of multi-atomic molecules using quantum phase estimation on IBM hardware. This problem is important in understating the optical interaction with molecules and is a good example of a classically intractable problem. As described by the authors of thispaper, accurate spectra of larger molecules have not been calculated.
The aim of this project is to implement the zero temperature approach in the paper above, and if there is time to include the modifications suggested in it and in this follow up paper. Our team has experience in quantum optics and molecular physics as well as a qualified qiskit developer.
Code, once public will be at this repo:
https://github.com/alexdzm/Qhack_open
If non-zero temperature systems will be considered, an extra register is required for phase estimation. More quits will help with solving this problem. More quits are useful even in the zero temperature case, where a larger register allows us to consider higher vibrational states that increase the accuracy of any simulation.
We want to use a real quantum device because understanding how real noise effects this algorithm will be important in understanding whether it will work on NISQ hardware, or if it will have to wait for more fault tolerant quantum hardware.
Jourdan
Testing noise-induced robustness in quantum classifiers
The capabilities of near-term quantum device are severely limited by the presence of experimental perturbation, thus a number of noise-mitigating approaches have been proposed. Yet, a slight amount of noise might be useful in some contexts: it's well-known that the careful addition of noise can ensure desirable properties such as differential privacy and robustness to adversarial examples [1,2,3].
Prior research in this direction investigated the properties of depolarizing noise, both theoretically and through numerical simulations [4]. In our project, we train a variational quantum classifier on the "California housing" dataset [5] and we test its robustness to adversarial examples crafted with the ART library [6].
We perform our experiments both on classical simulators (Pennylane, Qiskit) and on NISQ architectures (IBM, IonQ, Rigetti). Thus, we study for the first time the robustness properties ensured by realistic noise models.
References
[1] M. Lecuyer et al., Certified Robustness to Adversarial Examples with Differential Privacy, https://arxiv.org/abs/1802.03471
[2] L. Zhou and M. Ying, Differential Privacy in Quantum Computation, https://ieeexplore.ieee.org/document/8049724
[3] A. Angrisani, M. Doosti and Elham Kashefi, Differential Privacy Amplification in Quantum and Quantum-inspired algorithms, under review
[4] Y. Du et al., Quantum noise protects quantum classifiers against adversaries, https://arxiv.org/abs/2003.09416
[5] https://inria.github.io/scikit-learn-mooc/python_scripts/datasets_california_housing.html
[6] https://adversarial-robustness-toolbox.readthedocs.io/en/latest/#
https://github.com/Matx00/qhack-jourdan/blob/main/qhack.ipynb
Whereas prior work investigated differential privacy and robustness in quantum classifiers either theoretically or numerically, we conduct experiments on NISQ devices. Exploring the properties of the quantum noise that inherently affects the 16-qubit IBM machine would be extremely relevant for many use-cases.
qnyble
RNAs are nucleotide polymers fundamental to a diversity of elementary biological functions, including the (de)coding and regulation of genes, protein construction, cellular signaling, and catalysis [1]. Integral to these roles is the capacity and propensity of RNAs to self-interact through hydrogen-bond base-pairing between nucleotides and fold into specific, stable structures [2]. This folded structure of an RNA along with its primary sequence chemistry combine to dictate its interactions with other biomolecules [3]. Understanding and predicting RNA folding is thus a pressing interest of the biological sciences, basic and applied [4, 5].
RNA folding prediction from primary sequence information alone remains challenging classically, viewed from both the standpoints of chemical dynamics and combinatorial optimization of free energy. Quantum approaches, with their demonstrable advantage in both of these realms, therefore lend themselves well to the RNA folding problem. To our knowledge, only one attempt has been made to map RNA folding to quantum computing, via quantum annealing [6]. With this project, we seek to: (1) modify the Hamiltonian presented therein to better reflect the underlying chemistry of base-pairing, and (2) optimize the free parameters of the Hamiltonian against a suite of RNAs with known structures. Expanding on (2), we aim to implement and test our Hamiltonian with the quantum annealing hardware of D-Wave, and demonstrate a parallel approach with gate-based hardware via QAOA, using the very same Hamiltonian.
https://github.com/JuanGiraldo0212/Qhack-qnyble
To train the free parameters of our Hamiltonian to achieve the best possible RNA prediction, and understand more generally the efficacy of our model, we need to test over many structures, both with annealers and gate-based machines. On the annealing side, D-Wave affords public users one hour of QPU access time per month, and in our experience this is limiting to a project which requires many runs for calibration/optimization purposes.
If we were granted additional access to the D-Wave QPUs through AWS/Braket, we would be able to test over a more complete set of parameter combinations in trialling our solution, allowing us to better find and report limitations and improvement opportunities for this specific approach. Further, we would have the opportunity to work with the gate-based machines of IonQ and Rigetti in a similar manner (via QAOA as noted in the project description), which would permit us to compare results between machines and methods.
QuantumArtist
Travelling Salesman Problem (TSP) is a known NP-complete problem especially to computer scientists and mathematicians. The problem states that "There is a salesman who is travelling from one city to another to sell something. The goal is to find the shortest path between the cities so that the salesman will be able travel all the cities and return back in as much least time as possible".
Given the statement above, this project will aim to reduce the travel time of the salesman using an Quantum Approximate Optimization Algorithm (QAOA).
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
Given access to IBM 16-qubit QPU, I would use it to run the algorithm on deeper graphs to understand and eventually try to get higher optimization in the circuit and estimation values
While completing the challenges I noticed some of the python files have a #!
(shebang) line at the top to make them runnable from the command line. E.g. this file begins with #! /usr/bin/python3
. As far as I know, a more appropriate (and portable) shebang is #!/usr/bin/env python3
as env
is always in /usr/bin
, but python
isn't in that location on everyone's machines. Indeed running ./order_matters_template.py < 1.in
fails for me. More details can be found at this stack exchange answer.
Lastly, since the shebang is included in many files, I would expect them to be executable, however all the ones I've tested are not.
If this is something worthy of fixing, I can open a PR. Let me know what you think.
Avocados
The Quantum Counselor for portfolio investment is a tool with two main objectives: forecasting the trend of assets price and optimizing portfolio returns, both using quantum computing techniques. For the case of the forecasting method, we use a hybrid method that combines a deep learning model of classical LSTM layers with quantum layers. For the case of portfolio optimization, the quantum algorithms of QAOA and VQE are used to solve the problem and will be compared with CPLEX, a classical solver. Both tools are deeply connected because the forecasted price of the different assets is used for the cost function construction.
The source code of our project you can find here
We would like to have the IBM Quantum power-up because we want to test our small-size portfolio optimization QAOA model on real quantum devices to test error mitigation techniques. The Small size model consists of 5 stocks and 3 periods of time (this requires 15 qubits). We want to prove the best combination with different techniques for this project using VQE, QAOA, with various optimizers and do a comparative using the classical method Cplex.
anonymousr007
This project aims to use a truncated (regularized) Hamiltonian for the matrix quantum mechanics models. This Hamiltonian is constructed by considering a truncated Hilbert space in the Fock basis. The truncated Hilbert space is constructed starting from the individual matrix degrees of freedom.
Two types of matrix quantum mechanics models are used
Quantum mechanics with matrix degrees of freedom plays an important role in gauge-gravity duality. Gauge-gravity duality translates difficult problems in quantum gravity to well-defined problems in non-gravitational quantum theories. Although it originated from string–M-theory, connections to various other fields, including
We use the Variational Quantum EigenSolver (VQE) to estimate the low-energy spectrum As for the VQE, the specific architecture that we use does not show a satisfactory performance at strong coupling, perhaps due to the variational forms parametrized by the quantum circuits not adequately probing the full gauge-invariant Hilbert space. This result shows that going beyond the VQE and using more complicated or fully quantum algorithms is not the correct way to approach matrix quantum mechanics for now, because they would require even deeper quantum circuits that are more prone to noise on actual quantum hardware.
QHack 2022 Open Hackathon Project
Variational Quantum Eigensolvers (VQE) for calculating ground state energies of molecules are one of the major applications of noisy intermediate scale quantum (NISQ) computers. However for VQE to be viable on NISQ computers, powerful error mitigation protocols are needed due to the high level of noise.
In this project, we investigate applications of a learning based quantum error mitigation (LBEM) method [1] on VQE for molecular ground state energy calculation. LBEM models an error free result with a quasi probabilistic mixture of noisy results. This distribution is learned via an ab initio process, without prior knowledge on the hardware error model. Clifford circuits are used for the training, so classical simulation is efficient, and the mitigation takes account of both spatial and temporal correlations.
We would like to use a 16 qubit hardware to investigate the performance of LBEM on molecules beyond H2. Some candidate molecules are H2O and BeH2. LBEM is thought to be scalable to deep circuits on large systems.
QRiders
Solanaceous crops(Potato, Tomato, and Pepper) are one of the most cultivated and in-demand crops after rice and wheat. Solanaceous farming dominates as an occupation in the agriculture domain in more than 125 countries. However, even these crops are, subjected to infections and diseases, mostly categorized into two grades:
Farmers who grow these crops suffer from serious financial standpoint losses each year. Wouldn't it be amazing if the farmer could discover this ailment early and treat it properly? This can prevent a lot of waste and financial loss.
https://github.com/bopardikarsoham/QHack-Open-Hackathon
If I get access to AWS, I'll be able to explore the effect of QML algorithms on larger qubit devices. I will be able to encode larger images than a standard 2x2 or 4x4. This can significantly improve my project.
Your team's name (matching the name used on the QHack Coding Challenges, if applicable)
A brief description of your project (1-2 paragraphs).
Chinese Medicine has been used successfully in China to fight COVID-19 pandemic, and Chinese herbs are used together at the same time in treating diseases, including in fighting recent COVID-19. This project is to use currently available quantum computers to analyze various components in those herbs and their effects so that we’ll know what specific chemical substances/drugs in these herbs that we could use in fighting next pandemics of those viruses and their variants widely.
After my honorable graduation from Tongji Medical School at Huazhong University of Science and Technology where I learned Western Medicine and some Chinese Medicine, I did my Residency on Internal Medicine at Union (Xie He) Hospital. I then went to Germany and did my doctoral thesis at Munich University. After obtaining my doctoral degree with honor from Munich University, I came to University of California at Los Angeles (UCLA) for my post-doctoral fellowship and subsequently worked there. I also took and successfully passed the U.S. National Board Step-1, National Board Step-2 and Clinical Skill Assessment (CSA), and am certified by the U.S. Educational Commission for Foreign Medical Graduates (ECFMG).
In 2006, I moved from Los Angeles (UCLA) to Washington, DC for my job at the National Institutes at Health (NIH). My grandfather from my mother side had been a very good doctor practicing Chinese Medicine in Wuhan before he passed. My mother had been an excellent pediatrician who treated her patients with Western Medicine together with Chinese Medicine that she learned from her father. From this historical and family influence, I paid very close attention to how Chinese Medicine was successfully used in fighting COVID-19 in China since the outbreak in Wuhan, while I’ve been following up with COVID-19 in U.S.
Starting April 2020, I’ve been using Chinese Medicine on myself in fighting the challenges. My e-mail is [email protected] and my phone # is 240-453-1534. I look forward to hearing from you about my draft of this QHack Open Hackathon project.
Many thanks, Sincerely,
Yining Xie
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc)
https://doi.org/10.1145/3498691
Two Attachments here: 3498691.pdf; appendices.pdf
3498691.pdf
appendices.pdf
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
If awarded, the access to IBM Quantum machine with IBM 16-qubit QPU will be used to finish this project.
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc).
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
Hey there! I am José Ignacio Espinoza Camacho, a Master's student doing my research in quantum computing. I am taking part of the Coding Challenge under the team name JIEC.
Clustering is a set of mathematical and computational methods that are part of the unclassified learning techniques in Machine Learning. Clustering is frequently used to generate initial information from data sets about which little is known [1]. There are several families of algorithms. This work focuses on Spectral Clustering, specifically in Normalized Cuts [2]. In 2000, J. Shi and J. Malik designed the Normalized Cuts algorithm for image segmentation based on previous spectral clustering works. This algorithm holds an important characteristic: we can retrieve different segments of an image by using not only the second eigenvector (like usual spectral clustering algorithms), but a small set of eigenvectors.
This work lies amidst quantum machine learning and quantum image processing. Quantum Machine Learning (QML) is one of the fastest growing areas in quantum computing. In contrast with classical machine learning, QML finds atypical patterns more efficiently [3]. Quantum Image Processing is a relatively new area in quantum computing. This field focuses on storing, processing, and retrieving visual information (i.e. images and video) using quantum systems [4]. Based on the work of L. Tse, et al. [5], I pretend to explain and implement the QAOA algorithm they propose using Qiskit.
Here you will find the link to the work made by L. Tse, et al [5].
In this link you will find an explanatory jupyter notebook of my project Graph Cut Segmentation via QAOQ implemented with Qiskit
Finally, in the following link you will find the final source code of my project
Given the nature of the work [5], the Challenges I would like to apply are:
Access to 16-qubit IBM Quantum machine. The Dataset uses images of 4x4 pixels, each pixel is represented by 1 qubit. Hence, the algorithm would fit perfectly in the 16-qubit IBM Quantum machine.
IBM Qiskit Challenge, Sponsored by IBM Quantum and Université de Sherbrooke
Hybrid Algorithms Challenge, Sponsored by AQT. QAOA algorithms are Variational Algorithms, hence, they are also hybrid - quantum algorithms.
QAOA Challenge, sponsored by Entropica
[1] A. K. Jain, M. N. Murty, and P. J. Flynn. Data clustering: A review. ACM Comput. Surv., 31(3):264–323, September 1999.
[2] Jianbo Shi and J. Malik. Normalized cuts and image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(8):888–905, 2000.
[3] Jacob Biamonte, Peter Wittek, Nicola Pancotti, Patrick Rebentrost, NathanWiebe, and Seth Lloyd. Quantum machine learning.Nature, 549(7671):195–202,Sep 2017.
[4] Fei Yan and Salvador E. Venegas-Andraca.Quantum Image Processing. SpringerNature Singapore Pte Ltd., 2020.
[5] Lisa Tse, Peter Mountney, Paul Klein, and Simone Severini. Graph cut segmen-tation methods revisited with a quantum algorithm.CoRR, https://arxiv.org/abs/1812.03050.
CloudKite
In current NISQ devices, circuit noise hinders us from obtaining a satisfying outcome. Even for those quantum algorithms specifically designed for NISQ devices like variational quantum algorithms, noise could induce barren plateaus and affect the optimization performance[1].
A recent article proposed the idea for error mitigation using a noise-estimation circuit[2]: Before running the working circuit, an estimation circuit based on the working circuit will be executed and used to measure the noise scale. The author has shown that this idea could improve performance when simulating the Heisenberg model, and the improvement is even larger when circuit depth becomes larger. We are going to replicate the result using real noisy backends and apply the idea to more use cases like variational algorithms and QAOA.
[1] Wang, S., Fontana, E., Cerezo, M. et al. Noise-induced barren plateaus in variational quantum algorithms. Nat Commun 12, 6961 (2021).
[2] Urbanek, Miroslav, et al. "Mitigating depolarizing noise on quantum computers with noise-estimation circuits." Physical Review Letters 127.27 (2021): 270502.
The 16-qubit QPU could provide the chance to explore the idea's gain for quantum circuits with a large size, which allows us to verify the dominant range in terms of system(qubit) size for this error mitigation protocol. After we implemented the six-qubit example shown in ref [2], we will increase the system size step by step to a 16-qubit example. The number of total shots could be estimated by 16(steps)*3(one without error mitigation, two with error mitigation)*8192(shots) for each simulation.
The access of the 16-qubit device could also give the possibility for us to try more use cases. In general, variational algorithms with noise-estimation circuits would need twice the amount of original shots. In the optimization process, we estimate ~200 iteration to achieve significant performance improvement.
Team - alekospagon
The project is qualified for the 3 categories:
Project Description:
Abstract: The Project is an application of the QAOA algorithm on a KnapSack variant. Specifically, it is for a problem reducable to KnapSack for which we implemented a quantum circuit (with many optimizations). The problem was introduced to our team on IBM's Fall Challenge 2021. We took this problem further by implementing new mixers for faster convergence proposed in Quantum Optimization Heuristics with an Application to Knapsack Problems
Technical details: We transcribe the problem to the QAOA's C operator. We construct the circuit corresponding to it; then we propose 3 optimizations that reduce the circuit's depth and then we propose an optimization (new mixers) that improves the algorithm's convergence speed. After that, we define a metric for the circuit's accuracy and we measure it for different inputs and repetitions. Last, but not least, using our metric we demonstrate the clear advantage of the new mixers. We propose further optimizations.
We uploaded the Project Pdf, and one power point to present our work. The pdf contains all the details and the source code for the Qiskit Simulation.
QHACK___QAOA_KnapSack_Variants_with_Optimized_Mixers.pdf
QHACK___QAOA_KnapSack_Variants_with_Optimized_Mixers___Power_Point___Non_Technincal.pdf
AlFerjani ( registered as Qonlyme )
Quantum computers will tackle problems in different fields such as medical research and finance, where the protection of sensitive data is a must. But quantum computers on the cloud can be threaten for security, particulary when delegating a potential data to such computers. In this context, clients, with limited computational ability, will want to use the services offered by quantum computation and communication protocols, in a way that their privacy is guarantee.
Homomorphic encryption (HE) enables arbitrary computation on encrypted data without decryption. Similarly to classical HE, quantum homomorphic encryption (QHE) allows clients with limited computational ability to delegated computations to untrusted quantum servers.
This project, we developed QuHE, a simple library for quantum homomorphic enryption using Qiskit in which we implement some quantum homomorphic encryption protocols to allow a quantum computer to compute on encrypted data. Then, we provide an implementation of quantum search on encrypted data using QuHE.
https://github.com/FerjaniMY/QuHE
** More description will be given later before the final deadline of the hackathon.
We aim to use approximately 10 qubits on the IBM 16-qubit QPU, to test the implementation of quantum search on encrypted data on the real quantum hardware.
There is a typo in the docstring, the example operators do not commute. Change:
def join_operators(op1, op2):
"""This function will receive two operators that we know can be simplified
and returns the operator corresponding to the union of the two previous ones.
Args:
- op1 (list(str)): First Pauli word (list of Pauli operators), e.g., ["Y", "I", "Z", "I"].
- op2 (list(str)): Second Pauli word (list of Pauli operators), e.g., ["Y", "I", "X", "I"].
Returns:
- (list(str)): Pauli operator corresponding to the union of op1 and op2.
For the case above the output would be ["Y", "X", "Z", "I"]
"""
to
def join_operators(op1, op2):
"""This function will receive two operators that we know can be simplified
and returns the operator corresponding to the union of the two previous ones.
Args:
- op1 (list(str)): First Pauli word (list of Pauli operators), e.g., ["Y", "I", "Z", "I"].
- op2 (list(str)): Second Pauli word (list of Pauli operators), e.g., ["Y", "X", "I", "I"].
Returns:
- (list(str)): Pauli operator corresponding to the union of op1 and op2.
For the case above the output would be ["Y", "X", "Z", "I"]
"""
BladeRunner
We are working on a Quantum Optical Neural Network that mimics the Brain. We are incorporating Speech, Image and Memory data. We want to simulate the human brain, despite our limited knowledge of the human brain we can use GAN’s, CNN’s and Circuit Optimizing to simulate the connections between the speech to image and memory to motor senses.
We use photons to achieve this high speed of communication between neurons.
https://github.com/hrushikesh890/Qhack2022_BladeRunner
We will be using the AWS credits to run benchmarking tests against the AWS systems.
###Team Name:
QSchrodinger
Project Description:
genetic quantum algorithm (GQA). GQA is based on the concept and principles of quantum computing such as qubits and superposition of states. Instead of binary, numeric, or symbolic representation, by adopting qubit chromosome as a representation GQA can represent a linear superposition of solutions due to its probabilistic representation. As genetic operators, quantum gates are employed for the search of the best solution. Rapid convergence and good global search capability characterize the performance of GQA. The effectiveness and the applicability of GQA are demonstrated by experimental results on the knapsack problem, The results show that GQA is superior to other genetic algorithms using penalty functions, repair methods, and decoders.
a quantum genetic algorithm consist on:
1- First step, the algorithm begins preparing a superposition of all individuals, i.e., N, or chromosomes of population Q(t), Therefore, all individuals are represented by only one individual quantum register. That is, the entire population is represented by a single chromosome in a superposition state, One of the key steps of RQGA is the correlation between the individual quantum register |x>i and a fitness quantum register | f itness>i.
2- In a second step the algorithm searches for the maximum fitness. Once the operator F is applied, RQGA searches for the maximum fitness value based on the Grover’s search algorithm.
3- Finally, making a measure in the chromosome with maximum fitness is obtained.
Reference material:
https://www.mdpi.com/2073-431X/5/4/24/pdf
https://arxiv.org/pdf/cs/0403003.pdf
https://www.researchgate.net/publication/2573070_Genetic_Quantum_Algorithm_and_its_Application_to_Combinatorial_Optimization_Problem
###Source code:
https://github.com/echchallaouy/Quantum-genetics
###Resource Estimate:
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
We are going to use the IBM 16-qubit QPU, if awarded to simulate a large number of the population's chromosomes.
1- In the first place, Genes are the qubits.
2- Second, preparing the superposition of all individuals, or chromosomes of the population.
3- In third place, the oracle O marks the maximum fitness of |ψ>i, such that when the oracle is applied we obtain the superposition, this step is repeated a given number of iterations. The Grover’s maximum number of iterations is calculated as pi *sqrt(2**n)/4 where n is the number of qubits or length of the quantum chromosome.
4- In fourth and last place the Grover’s diffusion operator G finds the chromosome with a marked state.
5- Finally, making a measure, we get the state that points to chromosome with maximum fitness
How do proteins fold? One of the million-dollar questions that we haven’t known the answer for. Understanding why a particular protein only folds that way and why not the other hundreds of thousands of possibilities can take homo sapiens a long way. To invent new drugs and medicines, we need the protein’s structure and that is not that simple and easy. A moderate amino acid chain protein itself can have a conformation size of the mass of the Earth. With the increase in the amino acids in the chain, the potential conformations grow exponentially which is why this problem is hard to tackle using a classical computer.
Alzheimer's disease is a dreadful disease caused in humans where it destroys memory and other important mental functions. In India, over 1 million people suffer from Alzheimer’s disease every year. The main objective of this project is to use quantum computing and find the ground state of the BACE1 exosite binding peptide because almost all of these peptides fold into their ground states. We would love to use different techniques such as Error Mitigation to compare and check the results.
As far as this project is concerned, we have only taken the main chain with 6 amino acids and we require 6 qubits to run and execute this experiment. We have not included any side chain in this experiment because of the exponential rise in the number of qubits used and the time it takes to run. We would love to take this project further including the side chain which will obviously require access to a higher qubit machine with which we can find the ground state of the higher amino acid peptide and its conformation.
Amazon Bracket Challenge
IBM Qiskit Challenge
Bio-QML Challenge
Quantum Chemistry Challenge
BladeRunner
One of the most challenging and important tasks in chemistry is finding the stable geometry of molecules. Classically the problem is computationally intensive. Hence, there is huge interest in quantum computers to solve this problem. The problem can be formulated as an optimization problem, wherein we minimize the energy by placing the atoms in molecules properly. This can be achieved by modeling it using qubit Hamiltonian and solving it using VQE. VQE is the algorithm of choice as it has proven itself one of the best algorithms to run on current NISQ machines. However, is there a way we can make VQE even more efficient and resilient to noise?
The answer is yes! What if we could reduce the number of qubits required for the problem? This would translate to massive gains in saving computational cost and noise resiliency. In our project, we aim to do so using the qubit efficient encoding. We will contrast our method against the standard VQE method and demonstrate the better performance in presence of noise by running examples on a real quantum computer. The molecules we consider are H20 and H4 (2+) dication. Furthermore, we will also explore the extension of the method to dynamical systems.
Reference:
Shee, Yu, et al. "A Qubit-Efficient Encoding Scheme for Quantum Simulations of Electronic Structure." arXiv preprint arXiv:2110.04112 (2021).
https://github.com/yulunwang/QHack-OptimizeStructure/blob/main/README.md
We will run a toy example and demonstrate the effects of noise on standard VQE vs. qubit efficient VQE using hardware available in AWS.
ARE YA WINNING, SON?
Anastasiia Andriievska, Artem Kuzmichev
You certainly know the classical Uno game! Any card you put down must either be the same color or the same number as the card in the center.
https://github.com/artem-phys/quantum-uno
It is very important to evaluate how quantum circuits should be built on quantum computers with a large number of qubits. In a long perspective, our team planned to begin a start-up with implications on mitigating errors on quantum hardware. Gamification in this turn serves as an inspirational way to involve everyone to learn prominent quantum technologies.
Our project is qualified for:
UncertaintyHack
We implement a quantum-classical hybrid QLSTM model by incorporating quantum variational layers into the classical LSTM in order to improve the efficiency and trainability of LSTM for better stock price prediction.
Stock price prediction is one of the most rewarding problems in modern finance, where the accurate forecasting of future stock prices can yield significant profit and reduce the risks. LSTM (Long Short-Term Memory) is a recurrent Neural Network (RNN) applicable to a broad range of problems aiming to analyze or classify sequential data. LSTM can be used to predict the future stock price based on the historical data sequences. Recent studies have shown that its efficiency and trainability can be improved by leveraging a quantum-classical hybrid model of LSTM. QLSTM proved to learn significantly more information after the first training epoch than its classical counterpart. Thus, we implement a variational quantum-classical hybrid algorithm engaging machine learning techniques within the LSTM model framework in order to predict stock price movement.
Firstly, we conduct the feature preparation of the stock price using data collection of the technical indicators for a given stock, correlated assets, the sentiment analysis of the related sources of the information from the market: news, social media, reports. Then we test a classical multi-input LSTM on the prepared dataset and analyze the feasibility of this instrument to predict the prices. Finally, we modify classical LSTM by introducing variational quantum circuits and comparing the result to that of classical LSTM.
QLSTM shows better prediction accuracy (10 times less RMSE) and trainability compared to classical LSTM as shown by Fang et. al (2020) Furthermore, QLSTM requires a relatively low amount of qubits. The depth of the variational quantum circuit(d) grows linearly both on the number of variational layers and the number of qubits. An iterative optimization technique is used to update the variational parameters of the quantum circuit employed in the hidden layer of the Neural Network in a way, that each of the learned parameters can effectively absorb the surrounding noise without even knowing any properties of the noise. Low gate depth per run and lower qubit count, together with noise tolerance make the variational technique viable to use in NISQ. (Noisy Intermediate-Scale) era devices.
Please refer to our github repo here
For better trainability of LSTM, we need to try different numbers of variational layers, qubits, and circuit depth. As such, we will run more trials and use quantum computers on AWS with a greater number of qubits.
Samras
Quantum entanglement used as a resource adds an advantage that cannot be obtained with purely classical approaches. This phenomenon manifests itself in CHSH game and quantum superdense coding, where entanglement boosts the winning probabilities for the players and communication rate, respectively.
In our project, we investigate the use of entanglement in the task of quantum autoencoding. Quantum autoencoders [1] compress quantum information into smaller dimensional Hilbert spaces, and moreover, it is known that entanglement resources can aid in this compression [2]. We take state-of-the-art quantum autoencoder strategies, and add additional entanglement resources to test for a compression advantage. We aim to test the approach on datasets such that the known autoencoder schemes do not compress and decompress the data properly, but when we add the entanglement resources, the result improves. Indeed part of our project is to identify such datasets - at this point we are working with the breast cancer data set and with the credit card fraud dataset.
We believe that use-cases beyond current research directions might arise with regards to entanglement-assisted quantum autoencoders. They might play a role in quantum state transmission, where one party physically transmits a quantum state to another party. The two parties, as in superdense coding and the CHSH game, share entanglement ahead of time. The sending party uses her share of entanglement to aid in compressing the total state she wants to transmit and then transmits. The receiver uses his share to decompress the state and potentially de-noise his received quantum states. Such scenarios already arise in multi-processor quantum computing, where quantum states move between various quantum processors, or in schemes like quantum key distribution, where some schemes like E91 and DI-QKD, already deploy quantum entanglement as a resource.
[1] Romero, Jonathan, Jonathan P. Olson, and Alan Aspuru-Guzik. "Quantum autoencoders for efficient compression of quantum data." Quantum Science and Technology 2.4 (2017): 045001.
[2] Z. B. Khanian and A. Winter, "Entanglement-Assisted Quantum Data Compression," 2019 IEEE International Symposium on Information Theory (ISIT), 2019, pp. 1147-1151, doi: 10.1109/ISIT.2019.8849352.
https://github.com/VoicuTomut/temporaryRepoQhack
Firstly, we will estimate the potential of entangled-assisted autoencoders using quantum simulators. However, at last we aim to develop an approach suitable for NISQ devices and advantageous for state of the art quantum hardware tasks.
We intend to run experiments on QPUs for the validation or falsification of our concept. We will start with 4 qubits encoding MNIST dataset images using either angle (2 by 2 pixelated pictures) or amplitude encoding (4 by 4 pixelated pictures). The amount of entanglement qubits needed to gain an advantage is not yet known, as we are currently working towards preliminary results showing advantage. Later on we would like to work with breast cancer and finance datasets. For each we would need around 1000 backends runs with 2500 each. Our simulation results demonstrate that quantum auto encoders might serve as a great tool for anomaly detection for the above-mentioned cases. Furthermore, we aim to test and train our autoencoder approach using Qiskit's noisy simulation ability, and using the trained parameters, we will compare our results against the real hardware to get a good sense of how real noise influences our result.
Marpeq
Our project aims to implement the recent suggestion by Rahmani et al. to simulate the Laughlin wavefunction of the ⅓ FQH state. This suggestion takes advantage of a clever reduction of the 2D state to a 1D system. We plan to implement and execute a circuit, verify it yields the required quantum state, and measure interesting physical properties such as the mutual statistics of quasi-particles. Since current NISQ devices are small and scarcely available, we also intend to implement a pennylane version of the parallelization system proposed by Barrat. The system allows simulations of large systems using only a handful of qubits. Since the FQH state is topologically ordered, we cannot rely on measurements of a local order parameter to verify the circuit generates the intended state. Thus it necessitates the measurement of non-local string operators. This means we must extend Barrat’s algorithm to facilitate measurements of non-local operators. We believe this is possible because the FQH ground state assumes the form of a matrix product state.
One challenging issue we need to face in order to complete our plan is the mitigation of noise. Beyond the inherent noise that inflicts each qubit during each gate application Rahmani et al.’s suggestion assumes a snake-like connectivity topology of the qubits in order to use nearest-neighbor two-qubits. Since the device does not provide such connectivity, the noise is amplified by the requirement to execute additional SWAP gates. These allow the device to provide the connectivity between far away qubits. To overcome this issue, we will use simulator measurements of Rahmani et al.’s idealized circuit as training data for a variational circuit. The parameters of the variational circuit will be optimized to yield the FQH state.
Here is a link to our project’s GitHub repository, we note the work has only just begun:
https://github.com/ShapeshiftingMasterOfDarkness/QHack2022-FQHE
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
The circuit we will implement acts on 3-qubit blocks. Currently, we have access to IBM’s 7 qubit device, which allows maintaining the state of 2 qubit blocks concurrently. However, Rahmani et al. estimate 22 qubits as the desired number for suppression of boundary effects. Using the 16-qubit device will allow us to simulate more extensive systems, closer to that limit. Together with our parallelization scheme, we could accelerate the simulation and obtain more accurate and reliable measurements of more exciting and complicated physical processes.
For a system of N qubits simulated over d actual qubits, we will need to run O(N/d) measurement cycles. Since the circuit size is linear in the number d of qubits we expect the execution to be fast. So O(dN/d)=O(N) cycles in total are needed, thus we expect our use of the 16-qubit device to be efficient.
We note that Barratt et al. provided a qiskit implementation of their system. Therefore using it over the 16 qubit IBM device will save us a considerable amount of time.
qnyble
RNAs are nucleotide polymers fundamental to a diversity of elementary biological functions, including the (de)coding and regulation of genes, protein construction, cellular signaling, and catalysis [1]. Integral to these roles is the capacity and propensity of RNAs to self-interact through hydrogen-bond base-pairing between nucleotides and fold into specific, stable structures [2]. This folded structure of an RNA along with its primary sequence chemistry combine to dictate its interactions with other biomolecules [3]. Understanding and predicting RNA folding is thus a pressing interest of the biological sciences, basic and applied [4, 5].
RNA folding prediction from primary sequence information alone remains challenging classically, viewed from both the standpoints of chemical dynamics and combinatorial optimization of free energy. Quantum approaches, with their demonstrable advantage in both of these realms, therefore lend themselves well to the RNA folding problem. To our knowledge, only one attempt has been made to map RNA folding to quantum computing, via quantum annealing [6]. With this project, we seek to: (1) modify the Hamiltonian presented therein to better reflect the underlying chemistry of base-pairing, and (2) optimize the free parameters of the Hamiltonian against a suite of RNAs with known structures. Expanding on (2), we aim to implement and test our Hamiltonian with the quantum annealing hardware of D-Wave, and demonstrate a parallel approach with gate-based hardware via QAOA, using the very same Hamiltonian.
References:
[1] J. Li and C. Liu, “Coding or noncoding, the converging concepts of RNAs,” Frontiers in
Genetics, vol. 10, May 2019.
[2] G.L.ConnandD.E.Draper,“RNAstructure,”Current Opinion in Structural Biology,vol.8,
no. 3, pp. 278–285, Jun. 1998.
[3] S. R. Holbrook, “RNA structure: the long and the short of it,” Current Opinion in Structural
Biology, vol. 15, no. 3, pp. 302–308, Jun. 2005.
[4] M. D. Disney, “Targeting RNA with small molecules to capture opportunities at the intersec-
tion of chemistry, biology, and medicine,” Journal of the American Chemical Society, vol.
141, no. 17, pp. 6776–6790, Mar. 2019.
[5] N. G. Walter and L. E. Maquat, “Introduction—RNA: From single molecules to medicine,”
Chemical Reviews, vol. 118, no. 8, pp. 4117–4119, Apr. 2018.
[6] D. M. Fox, C. M. MacDermaid, A. M. Schreij, M. Zwierzyna, and R. C.
Walker, “RNA folding using quantum computers,” May 2021.
https://github.com/JuanGiraldo0212/Qhack-qnyble
Given that the proposed Hamiltonian for this specific problem works by mapping one qubit to a possible stem from a RNA sequence, we are directly constrained in the size of our inputs by the size of the selected quantum computer. Most RNAs have a large stem count making it impossible to use our methods to predict their secondary structure while only having access to the free-tier IBM machines. This would limit our experiments to small and trivial sequences which won't be able to give us complete feedback on how well our methodology is at solving the general problem.
If we were to have access to the IBM 16-qubit QPU we would be able to test over a more complete and expressive dataset of RNA sequences in trialling our solution, allowing us to find limitations and improvement opportunities for this specific approach.
Your team's name (matching the name used on the QHack Coding Challenges, if applicable)
A brief description of your project (1-2 paragraphs).
Severe acute respiratory syndrome coronavirus 2 (SARS CoV 2) for COVID-19, including its variants, has been wide spread globally. It affects all ages, races and various medical conditions. Using traditional computers is quite challenging to analyze interactions between proteome and genome, including gene-gene Interactions among virus and human genetics. This project is to use currently available quantum computers to analyze interactions between proteome and genome, including gene-gene interactions in SARS CoV 2 and human genetics.
Virology and genetics have been two fields that I’m very interested since my medical school. After my honorable graduation from Tongji Medical School at Huazhong University of Science and Technology where I learned Western Medicine and some Chinese Medicine, I did my Residency on Internal Medicine at Union (Xie He) Hospital. I then went to Germany and did my doctoral thesis at Munich University. After obtaining my doctoral degree with honor from Munich University, I came to University of California at Los Angeles (UCLA) for my post-doctoral fellowship and subsequently worked there. I also took and successfully passed the U.S. National Board Step-1, National Board Step-2 and Clinical Skill Assessment (CSA), and am certified by the U.S. Educational Commission for Foreign Medical Graduates (ECFMG).
In 2006, I moved from Los Angeles (UCLA) to Washington, DC for my job at the National Institutes at Health (NIH), and further strength my interests in Virology and genetics, particular at the time of current COVID-19 pandemics. My e-mail is [email protected] and my phone # is 240-453-1534. I am looking forward to hearing from you about my draft of this QHack Open Hackathon project.
Many thanks, Sincerely,
Yining Xie
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc).
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
https://doi.org/10.1145/3498691
Two Attachments here: 3498691.pdf; appendices.pdf
3498691.pdf
appendices.pdf
If awarded, the access to IBM Quantum machine with IBM 16-qubit QPU will be used to finish this project.
The Quantum Programming Language Twist from MIT will be used in conjunction with PenneLane (Pythone, Numby, PyTorch, TensorFlow, JAX interface, Cirq, Strawberrybields, Qiskitt, Forest, etc) as stated the above.
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
Qillers
Variational Quantum Eigensolver(VQE) is one of the most promising variational quantum/classical hybrid algorithms that efficiently find the minimum eigenvalue of a Hermitian matrix using near-term quantum computers. It is usually used in Quantum Chemistry to find the ground state energy of a simulated molecule. However, despite its success and promise as an algorithm to run over NISQ era quantum computers, it still faces some major challenges. One such challenge is to initialize good parameter heuristics that ensure rapid and consistent convergence to local minima of the parameterized quantum circuit landscape. This project uses the concept of the meta-learning(Classical Neural Network assisted VQEs), defined in this referenced paper[1], to rapidly find approximate optima in the parameter landscape. We train classical recurrent neural networks to find approximately optimal parameters within a small number of queries of the cost function for the VQE to demonstrate a significant improvement in the total number of optimization iterations required to reach a given accuracy. This project also aims to demonstrate various applications of the meta-learned VQE in Quantum Chemistry, Quantum Finance, and High Energy Physics and its potential speedup and performance improvement as compared to regular VQEs.
[1] Verdon, Guillaume, Mick Broughton, Jarrod R. McClean, Kevin J. Sung, Ryan Babbush, Zhang Jiang, Hartmut Neven and Masoud Mohseni. “Learning to learn with quantum neural networks via classical neural networks.” ArXiv abs/1907.05415 (2019): n. pag.
https://github.com/Siddharthgolecha/Qillers
I expect to use the IBM 16-qubit QPU, if awarded, to design the VQEs at their full potential by simulating larger molecules in Quantum Chemistry, and a more constrained and larger portfolio optimization in Quantum Finance. The QPU would also be beneficial in calculating the more precise ground state energy value of HEP particles.
Voyager
This project is a further study of Saesun Kim's research on applying QML in image classification (https://github.com/bagmk/Quantum_Machine_Learning_Express)
We are going to detect the animal in infrared camera and the car object in night camera by using QML and Grover algorithm.
After that , we are planning to compare the accuracy of detecting objects between those 2 methods.
Our goal is to apply QML and Grover algorithm, which was usually dealt only in theory, in practical fields like image segmentation and find the optimal method.
https://github.com/BrightSky77/Qhack_Quantum_Machine_Learning
By having access to AWS credits, we will be able to use more qubits so that we can process more big size images.
QuantumRing
Neutrino Oscillation which was first predicted by Bruno Pontecorvo in 1957, has been a great theoretical and experimental interest, as its precise properties can shed light on several properties of the neutrino. The experimental discovery of neutrino oscillation proves that neutrino has non-zero mass, which then causes a required modification to the Standard Model of particle physics. This experimental work by Takaaki Kajita and Arthur B. McDonald was so great that it was recognized with the 2015 Nobel Prize for Physics. Due to the potential of this process, many attempts have been conducted to gain a profound understanding of the phenomenon. In this project, we try to dive in and explore this complex quantum dynamic but with a different approach.
Using a quantum computer, our team try to simulate collective flavor oscillations which are created by the interaction neutrino-neutrino in a neutrino cloud with a high density of neutrinos. This process can happen in supernovae and the early universe - astrophysical scenarios with large neutrino density. In this project, we are considering a two-flavor case of interacting neutrinos which leads us to demonstrate the time and space evolution of the set of amplitudes from a Schrodinger equation:
The H - Hamiltonian from the equation is the Hamiltonian for neutrino flavor evolution in an environment with a high density of neutrinos which include vacuum and forward-scattering interaction contributions. Here we use QAOA (Quantum Approximate Optimization Algorithm) to realize this Hamiltonian with the ambition to scale the system to as many neutrinos as possible, this is where we need a quantum computer with many qubits to perform our quantum circuit. From this simulation, we aim to find a way to simulate many neutrinos interacting systems with polynomial scale-up which will become a great tool for researchers and scientists to look into this complex neutrino dynamic.
https://github.com/bachbao/Simulating-collective-neutrino-oscillation-using-QAOA-algorithm
In this project, we plan to use the power of a real quantum computer, therefore a great need for resources from a real quantum computer is what we want. By implementing the QAOA to realize the neutrino Hamiltonian, we will need to use quantum tomography which means running multiple circuits with many shots to retrieve the initial state. That's our first reason, we need resources from IBMQ to gain the ability to run many jobs on the IBMQ quantum computer. The second reason is in this current generation of quantum computers, noisy, near-term, intermediate size, it is important to check whether our method can work well using a real quantum computer as the project will lose most of its application meaning if it is only presented by using the simulator. This comes to our final and most important reason for demanding the resources from IBMQ power up, we plan to scale up our system to as many neutrinos as possible, therefore we want to have a large-scale quantum computer and the quantum computers from IBMQ quantum computer with 16 qubits - ibmq_guadalupe which can help us scale up to more than 10 neutrinos(really hard to achieve in classical).
Benjamin Hall, Alessandro Roggero, Alessandro Baroni, Joseph Carlson. "Simulation of Collective Neutrino Oscillations on a Quantum Computer." arXiv preprint arXiv:2102.12556 (2021).
Zewei Xiong. "Many-body effects of collective neutrino oscillations" arXiv preprint arXiv:2111.00437 (2021).
Your team's name (matching the name used on the QHack Coding Challenges, if applicable)
Quan
A brief description of your project (1-2 paragraphs).
Developing a quantum convolutional network with a skip connection network block.
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the additional AWS credits, if awarded, to finish your Open Hackathon project.
Mostly used in running Amazon Braket.
Estimating how to price Financial Derivatives - like options such as puts and calls - is a difficult task due to the huge number of possible changes in variables. While estimation techniques such as Classical Monte Carlo exist, they can easily rack up large 'error' or uncertainty; getting rid of this is time-consuming and costly.
In the case of the Classical Monte Carlo, for example, error scales with 1/sqrt(M) where M is the number of simulations. Because of this, in order to halve the error, you must quadruple the simulation number. To reduce the error to useful amounts, the quadratic scaling can mean large numbers of simulations are needed.
In Quantum Monte Carlo, however, we can offer a Quadratic Speedup, so error scales with 1/M. This has huge potential, since it can greatly improve the accuracy of Option Pricing while reducing the intensity of simulation required for them.
Team Quest hopes to explore this by implementing work in Quantum computational finance: Monte Carlo pricing of financial derivatives, seeing how Quantum Monte Carlo can be realised and executed.
Qumpula Quantum (I come from the University of Helsinki from Kumpula campus)
Data centers are consuming a huge amount of energy and producing an unbearable carbon footprint. This work proposes a data workflow optimization algorithm that divides the data workload among multiple data centers so that the computations can be performed as sustainably as possible.
Google has developed a carbon footprint metric that enables users to track their computations emissions. Inspired by the idea that in the future such detailed data would be available, the project proposes a solution how to divide the computations along with data centers (or even machines in the centers) so that the carbon footprint is minimized. The problem is formulated as a quadratic unconstrained binary optimization problem.
Since hackathons are also learning experiences, I planned to implement the algorithm using multiple different quantum hardware and software. This will include a D-wave's quantum annealers running in Amazon Braket and NISQ devices in Amazon Braket and in IBM Quantum systems.
See the full project description with references in this document which is along with the Github repo. I also planned to finalize the project as a demonstration system for a suitable database conference.
Source code, which is under work, is in Github repo Quantum Computing -based Optimization for Sustainable Data Workflows in Cloud Infrastructures.
I planned to scale the problem so that I would be able to demonstrate it on a 16-qubit machine.
BladeRunner
One of the most challenging and important tasks in chemistry is finding the stable geometry of molecules. Classically the problem is computationally intensive. Hence, there is huge interest in quantum computers to solve this problem. The problem can be formulated as an optimization problem, wherein we minimize the energy by placing the atoms in molecules properly. This can be achieved by modeling it using qubit Hamiltonian and solving it using VQE. VQE is the algorithm of choice as it has proven itself one of the best algorithms to run on current NISQ machines. However, is there a way we can make VQE even more efficient and resilient to noise?
The answer is yes! What if we could reduce the number of qubits required for the problem? This would translate to massive gains in saving computational cost and noise resiliency. In our project, we aim to do so using the qubit efficient encoding. We will contrast our method against the standard VQE method and demonstrate the better performance in presence of noise by running examples on a real quantum computer. The molecules we consider are H20 and H4 (2+) dication. Furthermore, we will also explore the extension of the method to dynamical systems.
Reference:
Shee, Yu, et al. "A Qubit-Efficient Encoding Scheme for Quantum Simulations of Electronic Structure." arXiv preprint arXiv:2110.04112 (2021).
https://github.com/yulunwang/QHack-OptimizeStructure/blob/main/README.md
We will run a toy example and demonstrate the effects of noise on standard VQE vs qubit efficient VQE on a 16-qubit IBM machine.
BladeRunner
A brief description of your project (1-2 paragraphs).
We are deciding to do a Quantum Computational Human Neural Network that mimics the human brain as given from BCI data. We will be using GAN's, QNN's, Autoencoders and LSTM's. We know that even in the biology field we still don't know everything about the human brain, but the synapses and connectivity of the parts such as the frontal, temporal, occipital and parietal lobes are key to basic functionality of life. We will be using Quantum Optical Neural Networks to achieve this, in this way we can use both pennylane and strawberry fields.
A hyperlink to the draft source code for your team's hackathon project (e.g., a GitHub repo).
https://github.com/hrahman12/QHack-2022-Open-Project-
A 1-2 paragraph written Resource Estimate, indicating how you expect to use the IBM 16-qubit QPU, if awarded, to finish your Open Hackathon project.
We will use it to benchmark cases for the Neural Network against other devices(both IBM and Amazon).
Qanything
The optimisation of molecular geometry is one of many fundamental problems in quantum chemistry. The motivation of this problem stem from the experimental observation that an optimal molecular geometry, which are found via numerical calculations, often correspond to actual molecular structure found in Nature. There is motivation in studying optimal geometry of molecules as sometimes, the unique properties of the substance are attributed to its special molecular structure. For example, the V shaped structure of water explains ice formation and open structure of ice crystals with lower density than liquid water.
In this project, we shall investigate performance of Problem-Inspired Ansatze in solving the optimisation problem of molecular geometry, which are built using Given rotations [1] as simple building blocks. Given rotations are particle-preserving variational circuits for which are useful for approximating molecular ground states. In particular, problem-inspired ansatze will be constructed based on the Exact Decomposition of Unitary Coupled Cluster Single and Double (UCCSD) Unitary [2] which are known to be notoriously difficult to implement in practice with the current quantum devices due to its need for deep circuits. Importantly, we plan to employ creative optimization strategies on the both simulated and real hardware for simple molecules, such as H2 and others up to LiH . We also wish to study how the noise can affect the final accuracy on the molecular geometry.
The smallest test case, Hydrogen (H2) molecule will require a quantum circuit at least 4 qubits with 15 independent pauli strings observables.
The largest test case, Lithium Hydride (LiH) molecule will require at least 12 qubits with 631 independent pauli strings observables.
[1] Arrazola, J. M., Matteo, O. D., Quesada, N., Jahangiri, S., Delgado, A., & Killoran, N. (2021). Universal quantum circuits for quantum chemistry.
[2] Evangelista, F. A., Chan, G. K.-L., & Scuseria, G. E. (2019). Exact parameterization of fermionic wave functions via unitary coupled cluster theory. The Journal of Chemical Physics, 151(24), 244112.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.