Uncertainty Quantification

13 public types in this namespace, organized by kind.

Models & Types (7)

Type Summary
BayesianNeuralNetwork<T> Implements a Bayesian Neural Network that provides uncertainty estimates with predictions.
ConformalClassifier<T> Implements Conformal Prediction for classification tasks.
DeepEnsemble<T> Implements Deep Ensembles for uncertainty estimation.
ExpectedCalibrationError<T> Computes the Expected Calibration Error (ECE) metric for evaluating probability calibration.
MCDropoutNeuralNetwork<T> Implements Monte Carlo Dropout for uncertainty estimation.
SplitConformalPredictor<T> Implements Split Conformal Prediction for regression tasks.
TemperatureScaling<T> Implements temperature scaling for probability calibration.

Layers (2)

Type Summary
BayesianDenseLayer<T> Implements a Bayesian dense (fully-connected) layer using variational inference.
MCDropoutLayer<T> Implements Monte Carlo Dropout layer for uncertainty estimation in neural networks.

Interfaces (2)

Type Summary
IBayesianLayer<T> Defines the contract for Bayesian neural network layers that support probabilistic inference.
IUncertaintyEstimator<T> Defines the contract for models that can estimate prediction uncertainty.

Options & Configuration (2)

Type Summary
BayesianNeuralNetworkOptions Configuration options for BayesianNeuralNetwork.
MCDropoutNeuralNetworkOptions Configuration options for MCDropoutNeuralNetwork.