Policy Gradient Methods
MADDPG
Introduced by Lowe et al. in Multi-Agent Actor-Critic for Mixed Cooperative-Competitive EnvironmentsMADDPG, or Multi-agent DDPG, extends DDPG into a multi-agent policy gradient algorithm where decentralized agents learn a centralized critic based on the observations and actions of all agents. It leads to learned policies that only use local information (i.e. their own observations) at execution time, does not assume a differentiable model of the environment dynamics or any particular structure on the communication method between agents, and is applicable not only to cooperative interaction but to competitive or mixed interaction involving both physical and communicative behavior. The critic is augmented with extra information about the policies of other agents, while the actor only has access to local information. After training is completed, only the local actors are used at execution phase, acting in a decentralized manner.
Source: Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Reinforcement Learning (RL) | 19 | 31.67% |
| Multi-agent Reinforcement Learning | 17 | 28.33% |
| Edge-computing | 3 | 5.00% |
| Management | 2 | 3.33% |
| Autonomous Driving | 2 | 3.33% |
| Decision Making | 2 | 3.33% |
| Face Recognition | 1 | 1.67% |
| Incremental Learning | 1 | 1.67% |
| Hierarchical Reinforcement Learning | 1 | 1.67% |
Adam
Batch Normalization
Convolution
Dense Connections
Experience Replay
ReLU
Weight Decay
