sentenai/reinforce

Issues

• Can't install the package

  #28 opened 4 years ago by onurcanbektas
  1

• Convert Q-learning (and/or Sarsa) into an example agent in reinforce-zoo

  #10 opened 7 years ago by stites
  2

• add a testable "convergence" criteria

  #11 opened 7 years ago by stites
  2

• break the dependency on the gym-http-api server -- call the gym directly

  #6 opened 8 years ago by stites
  8

• Push gym-http-api to hackage + stackage

  #20 opened 7 years ago by stites
  1

• factor out agents?

  #25 opened 6 years ago by KiaraGrouwstra
  1

• Add some policy gradient methods to the algorithms

  #13 opened 7 years ago by stites
  0

• [meta] Deep-RL work

  #14 opened 7 years ago by stites
  0

• Add a universe environment

  #15 opened 7 years ago by stites
  0

• Add Roboschool environment

  #16 opened 7 years ago by stites
  0

• Make 10-armed bandits n-armed bandits

  #18 opened 7 years ago by stites
  0

• Add to travis - weeder, hlint, brittany

  #21 opened 7 years ago by stites
  0

• Add python bindings

  #22 opened 7 years ago by stites
  0

• Migrate from RWST to a History typeclass

  #2 opened 8 years ago by stites
  0

• More Gym environments

  #1 opened 8 years ago by stites
  0

• Clean up and formalize State- and Action-Space typeclasses

  #3 opened 8 years ago by stites
  1

• Add documentation around why reset and observation have different data constructors

  #5 opened 8 years ago by stites
  0

• Render the haskell cartpole environments with SDL

  #9 opened 7 years ago by stites
  0

• Port Sutton's Acrobot code to haskell

  #8 opened 7 years ago by stites
  0

• Add eligibility trace variants in algorithms

  #12 opened 7 years ago by stites
  6

• Build docs with doc-server

  #7 opened 7 years ago by mitchellwrosen
  1