飞机游戏五 强化学习算法库

飞机游戏五 强化学习算法库

目录

一 stable-baseline3

介绍

利用stable-baselines3优雅便捷地玩转深度强化学习算法 - 知乎 (zhihu.com)

项目地址

GitHub - DLR-RM/stable-baselines3: PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

提供算法及适用场景

Implemented Algorithms

Name Recurrent Box Discrete MultiDiscrete MultiBinary Multi Processing
ARS1 ❌ ✔️ ✔️ ❌ ❌ ✔️
A2C ❌ ✔️ ✔️ ✔️ ✔️ ✔️
DDPG ❌ ✔️ ❌ ❌ ❌ ✔️
DQN ❌ ❌ ✔️ ❌ ❌ ✔️
HER ❌ ✔️ ✔️ ❌ ❌ ❌
PPO ❌ ✔️ ✔️ ✔️ ✔️ ✔️
QR-DQN1 ❌ ❌ ✔️ ❌ ❌ ✔️
RecurrentPPO1 ✔️ ✔️ ✔️ ✔️ ✔️ ✔️
SAC ❌ ✔️ ❌ ❌ ❌ ✔️
TD3 ❌ ✔️ ❌ ❌ ❌ ✔️
TQC1 ❌ ✔️ ❌ ❌ ❌ ✔️
TRPO1 ❌ ✔️ ✔️ ✔️ ✔️ ✔️
Maskable PPO1 ❌ ❌ ✔️ ✔️ ✔️ ✔️

1: Implemented in SB3 Contrib GitHub repository.

Actions gym.spaces:

  • Box: A N-dimensional box that containes every point in the action space.
  • Discrete: A list of possible actions, where each timestep only one of the actions can be used.
  • MultiDiscrete: A list of possible actions, where each timestep only one action of each discrete set can be used.
  • MultiBinary: A list of possible actions, where each timestep any of the actions can be used in any combination.

二 小雅 ElegantRL

介绍

小雅 ElegantRL: 基于PyTorch的轻量-高效-稳定的深度强化学习框架 - 知乎 (zhihu.com)

项目地址

GitHub - AI4Finance-Foundation/ElegantRL: Cloud-native deep reinforcement learning. 🔥

提供算法及适用场景

ElegantRL implements the following model-free deep reinforcement learning (DRL) algorithms:

DDPG, TD3, SAC, PPO, REDQ for continuous actions in single-agent environment,
DQN, Double DQN, D3QN for discrete actions in single-agent environment,
QMIX, VDN, MADDPG, MAPPO, MATD3 in multi-agent environment.

项目架构

image-20221208203017973

posted @ 2022-12-08 20:32  英飞  阅读(645)  评论(0)    收藏  举报