Actors

An Actor is the inference-time counterpart to a Worker. Where Workers run training pipelines in isolated subprocesses, Actors live inside the GUI process and are responsible for selecting actions during evaluation and interactive play, without starting any training loop.

Actor Documentation

• Actor Concept
  • What is an Actor?
  • The Two Actor Protocols
  • Data Containers
  • When to use Actors vs Workers
• Actor Architecture
  • High-Level Position in MOSAIC
  • ActorService Internals
  • Policy Mapping Integration
  • Placeholder Actors vs Real Actors
• Actor Lifecycle
  • Overview
  • Phase 1: Registration
  • Phase 2: Activation
  • Phase 3: Seed Broadcast
  • Phase 4: Episode Begin
  • Phase 5: Step Loop
  • Phase 6: Episode End
  • Phase 7: Session End
  • Error Handling
• Actor Types
  • HumanKeyboardActor
  • CleanRLWorkerActor
  • XuanCeWorkerActor
  • RayRLlibWorkerActor
  • Implementing a Custom Actor