IRLAgent<T>

Interfaces · AiDotNet.Interfaces

Marker interface for reinforcement learning agents that integrate with AiModelBuilder.

Implements: IFeatureAware, IFullModel<T, Vector<T>, Vector<T>>

Known implementations: ReinforcementLearningAgentBase<T>

For Beginners

An RL agent is just a special kind of model that learns through interaction with an environment. By implementing IFullModel, RL agents work with all of AiDotNet’s existing infrastructure:

  • They can be saved and loaded
  • They work with the AiModelBuilder pattern
  • They support serialization, cloning, etc.

The key difference is how they’re trained:

  • Regular models: trained on fixed datasets (x, y)
  • RL agents: trained by interacting with environments and getting rewards

How It Works

This interface extends IFullModel to ensure RL agents integrate seamlessly with AiDotNet’s existing architecture. RL agents are models where:

  • TInput = Tensor (state observations, though often flattened to Vector in practice)
  • TOutput = Vector (actions)

Methods

Method Summary
Dictionary<string, T> GetMetrics() Gets current training metrics.
T Train() Performs one training step using stored experiences.
Vector<T> SelectAction(Vector<T> state, bool explore) Selects an action given the current state observation.
void ResetEpisode() Resets episode-specific state (if any).
void StoreExperience(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool done) Stores an experience tuple for later learning.
void StoreExperience(Vector<T> state, Vector<T> action, T reward, Vector<T> nextState, bool terminated, bool truncated) Stores an experience tuple, telling termination apart from truncation.