MetaCollaborativeOptions<T, TInput, TOutput>

Options & Configuration ยท AiDotNet.MetaLearning.Options

Configuration options for Meta-Collaborative Learning across multiple task domains.

Inheritance: ModelOptions โ†’ MetaCollaborativeOptions<T, TInput, TOutput>

Implements: IMetaLearnerOptions<T>

How It Works

Meta-Collaborative Learning uses gradient alignment between concurrently adapted tasks to transfer cross-domain knowledge. Tasks with aligned gradient directions reinforce each other, while conflicting gradients are dampened. A domain-specific momentum buffer per task stabilizes cross-task transfer.

Constructors

Constructor Summary
MetaCollaborativeOptions(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.
bool EnableCheckpointing { get; set; }
bool UseFirstOrder { get; set; } Whether to use the first-order approximation of the meta-gradient.
double AlignmentWeight { get; set; } Weight for the gradient alignment loss between tasks.
double GradientMomentum { get; set; } Momentum coefficient for the domain-specific gradient buffer.
double InnerLearningRate { get; set; }
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 NumDomainSlots { get; set; } Number of domain slots for maintaining separate gradient histories.
int NumMetaIterations { get; set; }
int? RandomSeed { get; set; }

Methods

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