All pull requests

Pull requests

• Honor eval_epsilon=0.0 in the DQN eval policy

  #375·rootkiller6788 opened Sep 14, 2026·

  ·
  1

• Honor zero evaluation epsilon in DQN

  #374·sylvesterkaczmarek opened Sep 12, 2026·

  ·
  1

• Fix typo in end_of_episode_behavior error messages

  #371·irhyl opened Jun 29, 2026·

  ·
  1

• Implement make_policy in ARSBuilder

  #370·irhyl opened Jun 29, 2026·

  ·
  1

• Add learner tests for SAC and TD3 agents

  #369·irhyl opened Jun 29, 2026·

  ·
  1

• Add examples review agentic workflow and Copilot instructions

  #367·markevanrozeboom opened Mar 18, 2026·

  ·
  1

• fix(datasets): use get_from_first_device to safely de-shard tfds payload arrays

  #366·divyashreepathihalli opened Mar 5, 2026·

  ·

• Upgrade GitHub Actions for Node 24 compatibility

  #362·salmanmkc opened Jan 25, 2026·

  ·

• docs: clarify getting started entry points

  #361·Vinayaktoor opened Jan 16, 2026·

  ·

• Update Pytype to 2024.10.11 (Maintenance)

  #358·natinew77-creator opened Dec 20, 2025·

  ·

• Improve Error Messages for Distributional Losses (Fixes #308)

  #357·natinew77-creator opened Dec 20, 2025·

  ·

• Fix Dependencies, Improve Documentation, and Enhance Legal Actions

  #356·natinew77-creator opened Dec 20, 2025·

  ·

• fix: ensure CQL learner transitions from behavior cloning correctly

  #355·natinew77-creator opened Dec 11, 2025·

  ·
  1

• docs: Clarify installation instructions for running examples

  #354·natinew77-creator opened Dec 10, 2025·

  ·
  1

• fix: use launch_with_termination_handler for clean experiment termina…

  #353·natinew77-creator opened Dec 10, 2025·

  ·
  1

• Add utility for graceful Launchpad termination with external schedulers

  #352·natinew77-creator opened Dec 8, 2025·

  ·
  1

• Support nested observation structures in MCTS agent

  #351·natinew77-creator opened Dec 8, 2025·

  ·
  1

• Fix ImportError: Cannot import experiments from acme.jax

  #350·natinew77-creator opened Dec 7, 2025·

  ·
  1

• Fixes #347: Pin JAX/Chex dependencies in Colab notebooks

  #349·natinew77-creator opened Dec 7, 2025·

  ·
  1

  1

• Add Implicit Q-Learning (IQL) implementation

  #348·hwilner opened Nov 2, 2025·

  ·
  1

• Refactor: Agent Class

  #343·Vruddhi18 opened Jun 13, 2025·

  ·

• Improve README for clarity, structure, and onboarding

  #339·manipratapsingh1 opened Apr 5, 2025·

  ·

• Pinning dm-haiku to 0.0.10

  #313·jonas-eschmann opened Nov 17, 2023·

  ·
  1

• Add devices params to PPO learner

  #309·armandpicard opened Oct 23, 2023·

  ·

• Add samples per second logging for reverb_dataset.py

  #304·hartikainen opened Jun 20, 2023·

  ·