Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement RL algorithms #5

Open
3 of 8 tasks
Geonhee-LEE opened this issue Aug 21, 2020 · 5 comments
Open
3 of 8 tasks

Implement RL algorithms #5

Geonhee-LEE opened this issue Aug 21, 2020 · 5 comments
Assignees

Comments

@Geonhee-LEE
Copy link
Owner

Geonhee-LEE commented Aug 21, 2020

  • Value based RL

    • DQN
    • Rainbow DQN
    • CQL
  • Value based + Policy based RL

  • Policy based RL

    • PG
    • TRPO
    • PPO
  • Model based RL

    • [ ]
@Geonhee-LEE
Copy link
Owner Author

@CzJaewan 이전에 얘기했던 value based, policy based 분류는 https://spinningup.openai.com/en/latest/spinningup/rl_intro2.html#citations-below를 참조하면 좋을 듯.

  • 0825 진행사항) 지금은 개념다시 잡으면서 Everett training 진행 중 and 전통적인 물체 회피 빌드 및 테스트 진행 중.

ps) 보고바람

@CzJaewan
Copy link
Collaborator

@Geonhee-LEE value based RL, Policy based RL 등의 Algorithm들은 PPO를 통한 실제 테스트 까지 완료한 후 진행할 예정

이전 알고리즘 구현 및 테스트 진행 사항

  • Code 작성 완료 : DDPG, TD3, TRPO, SAC
  • Code Train 정상 작동 : DDPG
  • Code 수정 필요 : TD3, SAC
  • Code train test 필요 : TRPO

@Geonhee-LEE
Copy link
Owner Author

@CzJaewan 각각 소스의 참조 repo 좀 업데이트 부탁드립니다.

@Geonhee-LEE
Copy link
Owner Author

Geonhee-LEE commented Sep 24, 2020

SAC 구현

https://github.com/Geonhee-LEE/rl-collision-avoidance/tree/sac

  • argparse 추가
  • 참조한 SAC에 맞게 구조 변경
  • 현재 mpi4python으로 사용하기 위해 shape 맞춰주는 중. 참조
    image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants