Namespace AiDotNet.ReinforcementLearning.Policies
Classes
- BetaPolicyOptions<T>
Configuration options for Beta distribution policies.
- BetaPolicy<T>
Policy using Beta distribution for bounded continuous action spaces. Network outputs alpha and beta parameters for each action dimension. Actions are naturally bounded to [0, 1] and can be scaled to any [min, max] range.
- ContinuousPolicyOptions<T>
Configuration options for continuous action space policies in reinforcement learning. Continuous policies output actions as real-valued vectors using Gaussian (normal) distributions.
- ContinuousPolicy<T>
Policy for continuous action spaces using a neural network to output Gaussian parameters.
- DeterministicPolicyOptions<T>
Configuration options for deterministic policies.
- DeterministicPolicy<T>
Deterministic policy for continuous action spaces. Directly outputs actions without sampling from a distribution. Commonly used in DDPG, TD3, and other deterministic policy gradient methods.
- DiscretePolicyOptions<T>
Configuration options for discrete action space policies in reinforcement learning. Discrete policies select from a finite set of actions using categorical (softmax) distributions.
- DiscretePolicy<T>
Policy for discrete action spaces using a neural network to output action logits.
- MixedPolicyOptions<T>
Configuration options for mixed discrete and continuous policies.
- MixedPolicy<T>
Policy for environments with both discrete and continuous action spaces. Outputs both categorical distribution for discrete actions and Gaussian for continuous actions. Common in robotics where you have discrete mode selection and continuous parameter control.
- MultiModalPolicyOptions<T>
Configuration options for multi-modal mixture of Gaussians policies.
- MultiModalPolicy<T>
Multi-modal policy using mixture of Gaussians for complex action distributions.
- PolicyBase<T>
Abstract base class for policy implementations. Provides common functionality for numeric operations, random number generation, and resource management.
Interfaces
- IPolicy<T>
Core interface for RL policies - defines how to select actions.