Center for Human-Compatible AI ORG

Clean PyTorch implementations of imitation and reward learning algorithms

A benchmark environment for fully cooperative human-AI performance.

Reward Learning by Simulating the Past

Find best-response to a fixed policy in multi-agent RL

Library to compare and evaluate reward functions

Code for "On the Utility of Learning about Humans for Human-AI Coordination"

A prompt injection game to collect data for robust ML research

Web application where humans can play Overcooked with AI agents.

Dataset for the Tensor Trust project

Benchmark environments for reward modelling and imitation learning algorithms.

An Empirical Investigation of Representation Learning for Imitation (EIRLI), NeurIPS'21

Testing ranking algorithms to improve social cohesion

Reproduction of OpenAI and DeepMind's "Deep Reinforcement Learning from Human Preferences"

Code for "Evidence of Learned Look-Ahead in a Chess-Playing Neural Network"

No description.

(Experimental) Inverse reinforcement learning from trajectories generated by multiple agents with different (but correlated) rewards

Code accompanying "Learning What To Do by Simulating the Past", ICLR 2021.

Infer how suboptimal agents are suboptimal while planning, for example if they are hyperbolic time discounters.

No description.

Overcooked-AI Experiment Psiturk Demo (for MTurk experiments)

Training in bursts for defending against adversarial policies

Experiments in applying interpretability techniques to learned reward functions.

Highway driving simulation in JAX for Reinforcement Learning research

Checking the divisibility of neural networks, and investigating the nature of the pieces networks can be divided into.

Script for automatically creating the reconnaissance email.

Supporting code for Assistance Games as a Framework paper

Preprocessing reward functions to make them more interpretable

Code for the paper "Emergent Complexity via Multi-agent Competition"

Dataset for the paper: Political Neutrality as Balanced Approval: A Large-Scale Human Evaluation of AI Responses

No description.

PyTorch version of Stable Baselines, improved implementations of reinforcement learning algorithms.

OpenAI Gym wrapper for the DeepMind Control Suite

Supporting code for diagnostic seals paper

No description.

MineRL Competition for Sample Efficient Reinforcement Learning - Python Package

Prosocial Ranking Challenge Perspective Ranker

No description.

(experimental) RL agents should be more aligned if they do not know whether they are in simulation or in the real world

Sacred is a tool to help you configure, organize, log and reproduce experiments developed at IDSIA.

Use active learning to classify data represented as boundaries of regions in parameter space where a parametrised logical formula holds.

Iterative Linear Quadratic Regulator with auto-differentiatiable dynamics models

A training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.

LaTeX Notes from the Fall 2018 version of CS294-149: AGI Safety and Control

PRC: Testing ranking algorithms to improve social cohesion

PRC: Civirank submission

A fast and simple framework for building and running distributed applications. Ray is packaged with RLlib, a scalable reinforcement learning library, and Tune, a scalable hyperparameter tuning library.

A simple webpage that can visualize a sgf string encoded as a url fragment.

Docker files to help reproduce bug described in https://forums.developer.nvidia.com/t/kernel-oops-null-pointer-dereference-when-closing-cuda-application-katago/211270/3

Model summary in PyTorch similar to `model.summary()` in Keras

low disk space alerts posted to Slack

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:

A toolkit for developing and comparing reinforcement learning algorithms.

OpenAI Baselines: high-quality implementations of reinforcement learning algorithms

No description.

Traffic scenario definition and execution engine

Training framework for conditional imitation learning

Integration of AutoWare AV software with the CARLA simulator

No description.

No description.

No description.