Skip to content

Multi Agent Deep Reinforcement Learning consisting of Asset level agents, Domain level agents and a Master agent

Notifications You must be signed in to change notification settings

utsimul/SwanField

Repository files navigation

SwanField - Prototype version

alt text

Asset level PPO:

Shared Encoder (seq + non-seq) 
        ↓
    Shared Latent h
        ↓
   ┌───────────────┬───────────────┬───────────────┐
   │ Actor Head 1  │ Actor Head 2  │ Actor Head 3  │
   │ Categorical(3)│   Signal dist │ Memory dist   │
   └───────────────┴───────────────┴───────────────┘
                ↓
            Critic Head

About

Multi Agent Deep Reinforcement Learning consisting of Asset level agents, Domain level agents and a Master agent

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published