PEARLOptions<T, TInput, TOutput>

Options & Configuration · AiDotNet.MetaLearning.Options

Configuration options for PEARL: Probabilistic Embeddings for Actor-critic RL (Rakelly et al., ICML 2019).

Inheritance: ModelOptions → PEARLOptions<T, TInput, TOutput>

Implements: IMetaLearnerOptions<T>

How It Works

PEARL uses a probabilistic context encoder to infer a latent task variable z from transition data. The context encoder produces a Gaussian posterior q(z|c) that conditions the policy and value function. Task inference is amortized and gradient-free at test time — the encoder simply processes new transitions.

Constructors

Constructor Summary
PEARLOptions(IFullModel<T, TInput, TOutput> metaModel) Creates options for meta-learning metaModel, with the default settings.

Properties

Property Summary
IEpisodicDataLoader<T, TInput, TOutput> DataLoader { get; set; } Episodic data loader that supplies the meta-training tasks; null when tasks are passed to training directly.
IFullModel<T, TInput, TOutput> MetaModel { get; set; } The model whose parameters the meta-learner trains to adapt quickly to new tasks.
IGradientBasedOptimizer<T, TInput, TOutput> InnerOptimizer { get; set; } Optimizer for task adaptation.
IGradientBasedOptimizer<T, TInput, TOutput> MetaOptimizer { get; set; } Optimizer for the meta-update.
ILossFunction<T> LossFunction { get; set; } Loss function for meta-training; null uses the default regression loss.
bool EnableCheckpointing { get; set; }
bool UseFirstOrder { get; set; } Whether to use the first-order approximation of the meta-gradient.
double InnerLearningRate { get; set; }
double KLWeight { get; set; } KL divergence weight for the posterior regularization.
double OuterLearningRate { get; set; }
double? GradientClipThreshold { get; set; }
int AdaptationSteps { get; set; }
int CheckpointFrequency { get; set; }
int EncoderHiddenDim { get; set; } Context encoder hidden dimension.
int EvaluationFrequency { get; set; }
int EvaluationTasks { get; set; }
int LatentDim { get; set; } Dimensionality of the latent task variable z.
int MetaBatchSize { get; set; }
int NumMetaIterations { get; set; }
int NumPosteriorSamples { get; set; } Number of posterior samples for training.
int? RandomSeed { get; set; }

Methods

Method Summary
IMetaLearnerOptions<T> Clone()
bool IsValid()