epignatelli Goto Github PK

followers: 65.0 following: 11.0 repos: 128.0 gists: 19.0

Name: Eduardo Pignatelli

Type: User

Company: University College London (UCL)

Bio: Currently PhD candidate in RL at UCL. Previously ML Lead at @BuroHappoldEngineering and RA at @ImperialCollegeLondon

Twitter: EduPignatelli

Location: London

Blog: https://epignatelli.com

Hi traveller! 👋

I am Eduardo Pignatelli, I am an architect by education and a Machine Learning Scientist and Software Engineer by profession. I am interested in Deep Reinforcement Learning, its points of contact with neuropsychology, and its applications to the preservation of nature and the environment.

My trajectory

In my job positions I have:

Used Multi-Objective Genetic Algorithms to design the acoustics and the structure of human habitats as a Computational Architect
Helped design British stadia using machine learning as an Intern in the Stadia Atmosphere team at BuroHappold Engineering
Led the agile development of the UI/UX of the BHoM
Co-lead the design of the software architecture of the BHoM at BuroHappold Engineering
Led the Applied Machine Learning and Decision Making group at BuroHappold Engineering
Conducted research in Deep Learning and Computer Vision methods for the physiological modeling of cardiac tissue under arrythmya, at the Imperial College London
Started a PhD in Reinforcement Learning to study the Credit Assignment Problem (I know, I am late, I was busy), at UCL.

My hard skills

Programming is a second nature to me (C# / .NET, Python, C++, Javascript), as well as agile (and SCRUM) development. jax, pytorch, gym, and anything has got to do with RL and DL.
Data (structured, unstructured, online), databases (noSQL, please) and modelling (Reinforcement Learning, Deep Learning, Physics) yada yada..

My soft skills

I love working in groups and I work best in those, with a shared long-term objective and individual short-term responsibilities.
I am passionate, I do something if I believe in it. If I don't believe in it, I don't do it well. If you want the best out of me, tell me why the task is important, and I'll care for it, and dedicate my time, my mind to it.

My interests

I am interested in psychology (especially development psychology and neuropsychology), neuroscience and the mind (biological, synthetic, if any).
I teach -- I love teaching -- and I meander through the world -- I love the world -- often looking for quite, spiritual places far from the noise of the worldly life, seeking for that the ancient, lost craftmanships that every culture has.
My alter ego (one of many) is a cook -- and my send home is a kitchen.

You want to reach out because..

I want to hire you 🤞 https://epignatelli.com/cv
I want to collaborate on a project: edu dot pignatelli at gmail dot com
I want to challenge you at chess!

Eduardo Pignatelli's Projects

a-distributional-perspective-on-reinforcement-learning-marc

aaai-template

latex template for various conferences, as well as wise-man's overleaf (overleaf is terrible!)

aeolus

algorithms-basics

array_storage_benchmark

Compare some methods of array storage in Python (numpy)

at16k

Trained models for automatic speech recognition (ASR). A library to quickly build applications that require speech to text conversion.

autodidact

A pedagogical implementation of Autograd

automatic-watermark-detection

Project for Digital Image Processing

awesome-jax

JAX - A curated list of resources https://github.com/google/jax

bindsnet

Simulation of spiking neural networks (SNNs) using PyTorch.

boo

The Boo Programming Language.

bsuite

bsuite is a collection of carefully-designed experiments that investigate core capabilities of a reinforcement learning (RL) agent

bump-n-tag

A Github Action to automatically bump and tag master, on merge, with the latest SemVer formatted version. Works on any platform.

cardiax

A python implementation of the fenton karma model using fourth order accuracy central finite difference method, euler update scheme, and jax

cardiax_manuscript

ccxt

A JavaScript / Python / PHP cryptocurrency trading API with support for more than 120 bitcoin/altcoin exchanges

codeminion

A code generator framework capable of auto-generating the APIs of several SciSharp libraries.

community

Stores documents used by the TensorFlow developer community

continuous-minigrid

curated-rl-bibtex

A mantained collection of curated bibtex of the most relevant reinforcement learning publications

cv

deep-reinforcement-learning-with-double-q-learning

deep_bisim4control

Learning Invariant Representations for Reinforcement Learning without Reconstruction

deeplearningexamples

Deep Learning Examples

deq

desktop

Simple collaboration from your desktop

discovering-reinforcement-learning-algorithms

A Jax/Stax implementation of the general meta learning paper: Oh, J., Hessel, M., Czarnecki, W.M., Xu, Z., van Hasselt, H.P., Singh, S. and Silver, D., 2020. Discovering reinforcement learning algorithms. Advances in Neural Information Processing Systems, 33.