Avi Singh

maxtext. A simple, performant, and scalable Jax LLM!

perceiver-ar-pytorch. Implementation of Perceiver AR, Deepmind's new long-context attention network based on Perceiver architecture, in Pytorch

jaxrl. JAX (Flax) implementation of algorithms for Deep Reinforcement Learning with continuous action spaces.

BOReL. Official implementation for the paper "Offline Meta RL - Identifiability Challenges and Effective Data Collection Strategies", NeurIPS 2021.

nflows. Normalizing flows in PyTorch

ravens. Train robotic agents to learn pick and place with deep learning for vision-based manipulation in PyBullet. Transporter Nets, CoRL 2020.

tacto. Simulator of vision-based tactile sensors.

interbotix_ros_arms. ROS packages for the InterbotiX X-series family of robotic arms and turrets

robogym. Robotics Gym Environments

CQL. Code for conservative Q-learning

PlotNeuralNet. Latex code for making neural networks diagrams

BEAR. Code for Stabilizing Off-Policy RL via Bootstrapping Error Reduction

puppet. C

bullet3. Bullet Physics SDK: real-time collision detection and multi-physics simulation for VR, games, visual effects, robotics, machine learning etc.

replab. https://sites.google.com/view/replab/

BCQ. Author's PyTorch implementation of BCQ for continuous and discrete actions

state-marginal-matching. Efficient Exploration via State Marginal Matching (2019)

tensor2robot. Distributed machine learning infrastructure for large-scale robotics research

robosuite. robosuite: A Modular Simulation Framework and Benchmark for Robot Learning

notation. Collection of quotes on notation design & how it affects thought.

softlearning. Softlearning is a reinforcement learning framework for training maximum entropy policies in continuous domains. Includes the official implementation of the Soft Actor-Critic algorithm.

TD3. Author's PyTorch implementation of TD3 for OpenAI gym tasks

dm_control. Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.

garage. A toolkit for reproducible reinforcement learning research.

dopamine. Dopamine is a research framework for fast prototyping of reinforcement learning algorithms.

stable-baselines. A fork of OpenAI Baselines, implementations of reinforcement learning algorithms

evosoro. A Python Soft Robot Simulator Using Voxelyze

mujoco-py. MuJoCo is a physics engine for detailed, efficient rigid body simulations with contacts. mujoco-py allows using MuJoCo from Python 3.

supervised-reptile. Code for the paper "On First-Order Meta-Learning Algorithms"

rlkit. Collection of reinforcement learning algorithms

rl-teacher. Code for Deep RL from Human Preferences [Christiano et al]. Plus a webapp for collecting human feedback

spatial-reasoning. Code for the paper "Representation Learning for Grounded Spatial Reasoning"

tensorforce. Tensorforce: a TensorFlow library for applied reinforcement learning

dnc. A TensorFlow implementation of the Differentiable Neural Computer.

visdial. [CVPR 2017] Torch code for Visual Dialog

sonnet. TensorFlow-based neural network library

VIN_PyTorch_Visdom. PyTorch implementation of Value Iteration Networks (VIN): Clean, Simple and Modular. Visualization in Visdom.

homework. Assignments for CS294-112.

AirSim. Open source simulator for autonomous vehicles built on Unreal Engine / Unity, from Microsoft AI & Research

fold. Deep learning with dynamic computation graphs in TensorFlow

WassersteinGAN. Python

domain-transfer-network. TensorFlow Implementation of Unsupervised Cross-Domain Image Generation

tf-dann. Domain-Adversarial Neural Network in Tensorflow

predictron. Tensorflow implementation of "The Predictron: End-To-End Learning and Planning"

spatial-transformer-tensorflow. 🐝Tensorflow Implementation of Spatial Transformer Networks

ladder_network. Implementation of "Ladder Network" using Blocks and Theano.

imitation. Code for the paper "Generative Adversarial Imitation Learning"

simulated-unsupervised-tensorflow. TensorFlow implementation of "Learning from Simulated and Unsupervised Images through Adversarial Training"

research-advice. articles about research and phd advice

rodeo. A data science IDE for Python

semisupervised_vae. Replication of Semi-Supervised Learning with Deep Generative Models

auxiliary-deep-generative-models. Deep generative models for semi-supervised learning.

deer. DEEp Reinforcement learning framework

PixelDTGAN. A torch implementation of "Pixel-Level Domain Transfer"

visdial-amt-chat. [CVPR 2017] AMT chat interface code used to collect the Visual Dialog dataset

SGAN. Stacked Generative Adversarial Networks

universe-starter-agent. A starter agent that can solve a number of universe environments.

fast-weights. 🏃 Implementation of Using Fast Weights to Attend to the Recent Past.

readme2tex. Renders TeXy Math for Github Readme - No longer needed with official MathTex support on GH

PLSTM. Python

CoGAN. Jupyter Notebook

third_person_im. third person imitation learning. Archival only.

Inverse-Reinforcement-Learning. Implementations of selected inverse reinforcement learning algorithms.

GPflow. Gaussian processes in TensorFlow

ganhacks. starter from "How to Train a GAN?" at NIPS2016

MMD-critic. Terra

learning-to-learn. Learning to Learn in TensorFlow

universe. Universe: a software platform for measuring and training an AI's general intelligence across the world's supply of games, websites and other applications.

scipyplot. A Python Toolbox for Creating Scientific Article's Figures

neo. all open source hardware from comma.ai

openpilot. openpilot is an operating system for robotics. Currently, it upgrades the driver assistance system on 300+ supported cars.

gumbel-softmax. categorical variational autoencoder using the Gumbel-Softmax estimator

vae-tf. repository for the Variational Autoencoder (VAE) blogpost series from Fast Forward Labs

citysim3d. CitySim3D: Simulated car following benchmark

pix2pix. Image-to-image translation with conditional adversarial nets

Retro-Learning-Environment. The Retro Learning Environment (RLE) -- a learning framework for AI

gym_torcs. C++

vime. Code for the paper "Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks"

research-group-resources. Resources for people running research groups.

awesome-malmo-minecraft.

DailyCalHousingAnalysis. Jupyter Notebook

malmo. Project Malmo is a platform for Artificial Intelligence experimentation and research built on top of Minecraft. We aim to inspire a new generation of research into challenging new problems presented by this unique environment. --- For installation instructions, scroll down to *Getting Started* below, or visit the project page for more information:

keras-js. Run Keras models in the browser, with GPU support using WebGL

cartographer. Cartographer is a system that provides real-time simultaneous localization and mapping (SLAM) in 2D and 3D across multiple platforms and sensor configurations.

WaveFunctionCollapse. Bitmap & tilemap generation from a single example with the help of ideas from quantum mechanics

nips14-ssl. Code for reproducing results of NIPS 2014 paper "Semi-Supervised Learning with Deep Generative Models"

iGAN. Interactive Image Generation via Generative Adversarial Networks

fast-wavenet. Speedy Wavenet generation using dynamic programming :zap:

ladder. Ladder network is a deep learning algorithm that combines supervised and unsupervised learning.

cnn-benchmarks. Benchmarks for popular CNN models

videogan. Generating Videos with Scene Dynamics. NIPS 2016.

deeplearning-papernotes. Summaries and notes on Deep Learning research papers

keras-finetuning. Training an InceptionV3-based image classifier with your own dataset

paper-notes. Random notes on papers, likely a short-term repo.

research. dataset and code for 2016 paper "Learning a Driving Simulator"

betago. BetaGo: AlphaGo for the masses, live on GitHub.

keras-rl. Deep Reinforcement Learning for Keras.

modular_rl. Implementation of TRPO and related algorithms

DeepLearningVideoGames. Python

bitcoinbook. Mastering Bitcoin 3rd Edition - Programming the Open Blockchain