quantification

Quantification methods for uncertainty.

Submodules

decomposition

Uncertainty decomposition methods.

measure

Measures the uncertainty of a given representation.

notion

Definition of different notions of uncertainty.

Classes

QuantificationResult

Protocol for uncertainty quantifications.

Quantifier

Protocol for uncertainty quantification methods.

Functions

conditional_entropy

Compute the conditional entropy of a distribution.

conditional_spectral_entropy

Compute the mean spectral entropy within groups of embedding samples [WGD+26].

conformal_set_size

Measure conformal-set uncertainty by set size.

decompose

Decomposes the uncertainty of a given representation.

dempster_shafer_uncertainty

Compute the Dempster-Shafer uncertainty score of a distribution over logits.

entropy

Compute the entropy of a distribution.

entropy_of_expected_predictive_distribution

Compute the entropy of the expected value of a second-order distribution.

expected_max_probability_complement

Compute the expected value of one minus the max probability under a second-order distribution.

max_disagreement

Compute the expected gap between each sample's max probability and its probability on the BMA argmax.

max_probability_complement_of_expected

Compute one minus the max probability of the expected value of a second-order distribution.

mean_squared_distance_to_scaled_one_hot

Mean over members of \(\| h_k - s\, e_{\arg\max_j h_{k,j}} \|_2^2\); class axis is last.

measure

Measures the uncertainty of a given representation.

measure_atomic

Measures the uncertainty of a given representation without attempting to decompose it.

min_expected_total_variation

Compute the distance-based epistemic uncertainty of a second-order distribution.

mutual_information

Compute the mutual information of a distribution.

notion_registry

A simple switch-case mechanism based on a dictionary.

quantify

Quantifies the uncertainty of a given representation.

sample_variance

Measure uncertainty for samples via their sample variance.

spectral_entropy

Compute spectral uncertainty as Von Neumann entropy of an RBF kernel [WGD+26].

total_logit_sample_variance

Measure uncertainty for samples via the variance of their total logits (logits summed across members).

vacuity

Compute the vacuity of a second-order distribution.