AutoDiffTabOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for AutoDiff-Tab, an automated diffusion model for tabular data that searches over diffusion configurations and noise schedules to find optimal settings.

Inheritance: ModelOptions → NeuralNetworkOptions → ModelHyperparameterOptions → FinancialNeuralNetworkOptions → RiskModelOptions<T> → SyntheticDataGeneratorOptions<T> → AutoDiffTabOptions<T>

For Beginners

AutoDiff-Tab is like TabDDPM with automatic tuning:

Instead of manually setting “how noisy” and “how many steps” the diffusion process uses, AutoDiff-Tab tries different settings and picks the best one automatically.

It explores:

  1. Different amounts of noise (noise schedule)
  2. Different numbers of denoising steps
  3. Different neural network sizes

Example:

How It Works

AutoDiff-Tab automates the design of diffusion models for tabular data by searching over:

  • Number of diffusion timesteps: How many denoising steps
  • Beta schedule: Linear, cosine, or learned noise schedule
  • MLP architecture: Depth, width, and dropout of the denoiser
  • Noise schedule parameters: Start/end beta values

Reference: “Automated Diffusion Models for Tabular Data” (2024)

Constructors

Constructor Summary
AutoDiffTabOptions()

Properties

Property Summary
BetaSchedule BetaSchedule { get; set; } Gets or sets the type of noise schedule.
double BetaEnd { get; set; } Gets or sets the maximum beta value for the noise schedule.
double BetaStart { get; set; } Gets or sets the minimum beta value for the noise schedule.
double DropoutRate { get; set; } Gets or sets the dropout rate for the denoiser.
double LearningRate { get; set; } Gets or sets the learning rate.
int BatchSize { get; set; } Gets or sets the training batch size.
int Epochs { get; set; } Gets or sets the number of training epochs.
int MaxTimesteps { get; set; } Gets or sets the maximum number of diffusion timesteps to search over.
int SearchTrials { get; set; } Gets or sets the number of search trials to find optimal diffusion configuration.
int TimestepEmbeddingDimension { get; set; } Gets or sets the dimension of timestep embeddings.
int TrialEpochs { get; set; } Gets or sets the number of epochs for each search trial (reduced for speed).
int VGMModes { get; set; } Gets or sets the number of VGM modes for continuous column transformation.
int[] MLPDimensions { get; set; } Gets or sets the hidden layer dimensions for the denoiser MLP.