Skip to content

v0.2.0 - baseline reinforcement learning algorithms

No due date 0% complete
  • (envs) Possibly split reinforce into third-party dependencies (ie: -envs-gym, -envs-atari)
  • (envs) Introduce some kind of History monad to cleanly seperate debug logging from agent or environment reporting
  • (algos) write out more baseline algorithms in reinforce-algorithms (in progress)
  • (algos) modify project structure to so that reinforce and reinforce-al…
  • (envs) Possibly split reinforce into third-party dependencies (ie: -envs-gym, -envs-atari)
  • (envs) Introduce some kind of History monad to cleanly seperate debug logging from agent or environment reporting
  • (algos) write out more baseline algorithms in reinforce-algorithms (in progress)
  • (algos) modify project structure to so that reinforce and reinforce-algorithms are siblings

No hackage release with this. This would basically prototype a reinforce-algorithms project.

Loading