MetaDDPMOptions<T, TInput, TOutput>

Options & Configuration ยท AiDotNet.MetaLearning.Options

Configuration options for Meta-DDPM: Meta-Learning with Denoising Diffusion Probabilistic Models.

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

Implements: IMetaLearnerOptions<T>

How It Works

Meta-DDPM extends the DDPM framework to meta-learning by meta-learning a noise prediction network that generates task-specific model weights conditioned on support set embeddings. Unlike MetaDiff (which models weight deltas), Meta-DDPM directly generates the full adapted weight vector using the DDPM generative framework with a learned linear noise schedule and task-conditional denoising.

Constructors

Constructor Summary
MetaDDPMOptions(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 BetaEnd { get; set; } Ending beta for the linear noise schedule.
double BetaStart { get; set; } Starting beta for the linear noise schedule.
double EmaDecay { get; set; } Weight for EMA (exponential moving average) of model parameters for stable generation.
double InnerLearningRate { get; set; }
double OuterLearningRate { get; set; } Learning rate of the outer (meta) update.
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 NumTimesteps { get; set; } Total timesteps in the DDPM diffusion process.
int SamplingSteps { get; set; } Number of denoising steps used during generation (can be less than NumTimesteps).
int TaskConditionDim { get; set; } Dimensionality of the task conditioning vector.
int? RandomSeed { get; set; }

Methods

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