jaks19 Goto Github PK

1806

18.06 course at MIT in Spring 2017

abc-music-player

Software parses musical pieces in abc notation (.abc files) and plays them

adatest

Find and fix bugs in natural language machine learning models using adaptive testing.

adversarial-examples

augmented-reality-google-glass-safari-game

An augmented reality that places cute running animals all around your room. Gotta photograph 'em all!

bipedal-walker-adaptable-morphology

BipedalWalker-V2 on varying terrains, with morphology adaptation

checkers-game

An ongoing Web Application project which allows a player to play Checkers against the computer. Rules and graphics have been properly implemented and the game works with limited functionality. Final update on user input manipulation up and coming. (Right now a user can move by requesting any legal move but now user will be allowed to click on a checker and move it instead).

cnn_sentence_classification

pytorch Convolutional Networks for Sentence Classification - http://www.aclweb.org/anthology/D14-1181

cyclegan-pytorch

A very simple implementation of cyclegan, which is based on pytorch.

data_dump_for_ui

deeplearning.ai

deeplearning.ai , By Andrew Ng, All slide and notebook + code and some material.

distiller

Neural Network Distiller by Intel AI Lab: a Python package for neural network compression research. https://nervanasystems.github.io/distiller

ensemble_gms

Undirected graphical models are compact representations of joint probability distributions over random variables. To solve inference tasks of interest, graphical models of arbitrary topology can be trained using empirical risk minimization. However, to solve inference tasks that were not seen during training, these models (EGMs) often need to be re-trained. Instead, we propose an inference-agnostic adversarial training framework which produces an infinitely-large ensemble of graphical models (AGMs). The ensemble is optimized to generate data within the GAN framework, and inference is performed using a finite subset of these models. AGMs perform comparably with EGMs on inference tasks that the latter were specifically optimized for. Most importantly, AGMs show significantly better generalization to unseen inference tasks compared to EGMs, as well as deep neural architectures like GibbsNet and VAEAC which allow arbitrary conditioning. Finally, AGMs allow fast data sampling, competitive with Gibbs sampling from EGMs.

evolving-robotic-gripper

Optimizing the morphology of robotic fingers for increased object grasping accuracy on adversarially-shaped objects

itranslate

Web application allows English speakers learning Spanish to practice their skills by giving the correct English translation of Spanish words.

multiplayer-minesweeper

Hosts a Server to play command-line Multiplayer Minesweeper. Objective of the game: Stay alive for the longest possible time.

named-entity-recognition

Max entropy and RNN models used to detect GENE NAMES in biological academic literature

nlp-question-retrieval-lstm-cnn

nlp-transfer-learning-domain-adaptation

Using LSTM and CNN trained on the stack exchange Ubuntu dataset for the question similarity task on the Android dataset, using transfer learning for domain adaptation.

nnets-klines-partialmentalstates

Repo of code used in my 6.UAP Undergraduate EECS Thesis at MIT

parallel_gripper_simulation_pybullet

Parallel WSG-32 Gripper Simulation with a GUI wrapper and a Grasping-Routine wrapper

pytorch-deep-learning

Implementations of Machine Learning Algorithms in PyTorch

save-princess-cubed

A Unity3D game where the user is a cube that has to navigate through a map full of obstacles and mazes to reach its goal and keep moving up the levels.

scene-generation-from-novel-viewpoints-gens

Scene generation from novel viewpoints - Graph element networks

sentiment-analysis-nn-embeddings

A neural net model used to classify movie review sentiments using word embeddings for reviews

sim_jaco_robotiq

A URDF of a Jaco2 arm and Robotiq 2-finger gripper, and an example PyBullet environment for control

studyusa-web-portal

A portal for Students and Admin-alike to manage subscriptions, payments, file submission and review and aims to build community through a forum.