Skip to content
@HumanCompatibleAI

Center for Human-Compatible AI

CHAI seeks to develop the conceptual and technical wherewithal to reorient the general thrust of AI research towards provably beneficial systems.

Pinned Loading

  1. imitation imitation Public

    Clean PyTorch implementations of imitation and reward learning algorithms

    Python 1.2k 230

  2. overcooked_ai overcooked_ai Public

    A benchmark environment for fully cooperative human-AI performance.

    Jupyter Notebook 656 137

  3. rlsp rlsp Public

    Reward Learning by Simulating the Past

    Python 43 6

  4. adversarial-policies adversarial-policies Public

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

    Python 269 47

  5. evaluating-rewards evaluating-rewards Public

    Library to compare and evaluate reward functions

    Python 61 7

  6. human_aware_rl human_aware_rl Public

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

    Python 107 45

Repositories

Showing 10 of 55 repositories
  • ranking-challenge Public

    Testing ranking algorithms to improve social cohesion

    HumanCompatibleAI/ranking-challenge’s past year of commit activity
    Python 25 3 1 0 Updated Jun 29, 2024
  • overcooked_ai Public

    A benchmark environment for fully cooperative human-AI performance.

    HumanCompatibleAI/overcooked_ai’s past year of commit activity
    Jupyter Notebook 656 MIT 137 4 0 Updated Jun 25, 2024
  • leela-interp Public

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

    HumanCompatibleAI/leela-interp’s past year of commit activity
    Jupyter Notebook 9 GPL-3.0 0 0 0 Updated Jun 4, 2024
  • imitation Public

    Clean PyTorch implementations of imitation and reward learning algorithms

    HumanCompatibleAI/imitation’s past year of commit activity
    Python 1,207 MIT 230 67 19 Updated May 28, 2024
  • tensor-trust Public

    A prompt injection game to collect data for robust ML research

    HumanCompatibleAI/tensor-trust’s past year of commit activity
    Python 37 BSD-2-Clause 5 32 2 Updated Mar 21, 2024
  • tensor-trust-data Public

    Dataset for the Tensor Trust project

    HumanCompatibleAI/tensor-trust-data’s past year of commit activity
    Jupyter Notebook 29 2 0 0 Updated Mar 17, 2024
  • reward-function-interpretability Public
    HumanCompatibleAI/reward-function-interpretability’s past year of commit activity
    Jupyter Notebook 1 0 4 0 Updated Nov 30, 2023
  • evaluating-rewards Public

    Library to compare and evaluate reward functions

    HumanCompatibleAI/evaluating-rewards’s past year of commit activity
    Python 61 Apache-2.0 7 4 2 Updated Oct 23, 2023
  • seals Public

    Benchmark environments for reward modelling and imitation learning algorithms.

    HumanCompatibleAI/seals’s past year of commit activity
    Python 43 MIT 6 6 1 Updated Sep 19, 2023
  • human_aware_rl Public

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

    HumanCompatibleAI/human_aware_rl’s past year of commit activity
    Python 107 45 0 0 Updated Apr 17, 2023
View all repositories

Top languages

Loading…

Most used topics

Loading…