All pull requests

Pull requests

• Create a copy of agent in evaluation for chainerrl/experiements/train_agent_async.py

  #613·CyanFi opened Aug 10, 2021·

  ·

• FQF (Fully Parameterized Quantile Function) agent

  #579·muupan opened Nov 10, 2019Member·1 task·

  ·
  1

• Support 'spawn' in async training

  #574·muupan opened Nov 7, 2019Member·

  ·
  1

• [WIP] Use chainer.as_variable

  enhancement
  #510·muupan opened Jul 18, 2019Member·

  ·
  3

• [WIP] IMPALA-style actor-learner parallelism for DQN variants

  #477·muupan opened Jun 11, 2019Member·

  ·

• [WIP] Support chainer.distributions in TRPO

  #454·muupan opened May 6, 2019Member·

  ·

• [WIP] ChainerX support by DQN

  #375·muupan opened Dec 28, 2018Member·2 tasks·

  ·

• [WIP] Implements Hindsight Experience Replay

  #361·prabhatnagarajan opened Nov 26, 2018Contributor·

  ·
  4

  

• add act_with_exploration() method

  #298·LorenzoM1997 opened Aug 17, 2018·

  ·
  1

  

• Rewrite PCL agent

  #245·lyx-x opened Feb 25, 2018Contributor·

  ·
  1

• [WIP] Avoid action_space.sample

  #244·muupan opened Feb 15, 2018Member·

  ·
  1

• Fix the usage of t_max in PCL

  #237·lyx-x opened Feb 14, 2018Contributor·

  ·

• [WIP] Enable appending arbitrary values to replay buffer

  #224·muupan opened Jan 29, 2018Member·

  ·

• [WIP] Save and load lists

  #223·muupan opened Jan 29, 2018Member·

  ·