MetaTaskOptions<T, TInput, TOutput>

Options & Configuration · AiDotNet.MetaLearning.Options

Configuration options for the MetaTask algorithm, an implementation of MetaMix (Chen et al., ICPR 2020).

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

Implements: IMetaLearnerOptions<T>

How It Works

MetaMix keeps the MAML inner loop on the support set and replaces each episode’s query set by virtual examples mixed from random pairs of examples with a Beta-distributed coefficient.

Constructors

Constructor Summary
MetaTaskOptions(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 multi-class classification loss.
MetaMixSource MixSource { get; set; } Which set the two mixed examples are drawn from.
bool EnableCheckpointing { get; set; }
bool UseFirstOrder { get; set; } Whether to use the first-order approximation of the meta-gradient.
double InnerLearningRate { get; set; }
double InterpolationAlpha { get; set; } Alpha parameter of the symmetric Beta(α, α) distribution that draws the mixup coefficient λ (MetaMix Eq.
double OuterLearningRate { get; set; }
double? GradientClipThreshold { get; set; }
int AdaptationSteps { get; set; }
int CheckpointFrequency { get; set; }
int EvaluationFrequency { get; set; }
int EvaluationTasks { get; set; }
int MetaBatchSize { get; set; }
int NumMetaIterations { get; set; }
int NumMixedPairs { get; set; } Number Z of virtual (mixed) examples generated per episode (MetaMix Algorithm 1, line 14).
int? RandomSeed { get; set; }

Methods

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