shapiq.approximator¶

Algorithms for approximating Shapley values and interaction indices.

All approximators inherit from Approximator and follow a common interface: pass a Game and a computational budget, and receive InteractionValues as output.

Classes

`Approximator`	This class is the base class for all approximators.
`PermutationSamplingSII`	Permutation Sampling approximator for the SII (and k-SII) index.
`PermutationSamplingSTII`	Permutation Sampling approximator for the Shapley Taylor Index (STII).
`PermutationSamplingSV`	The Permutation Sampling algorithm for estimating the Shapley values.
`StratifiedSamplingSV`	The Stratified Sampling algorithm for estimating the Shapley values.
`OwenSamplingSV`	Owen Sampling approximator for the Shapley values.
`KernelSHAP`	The KernelSHAP regression approximator for estimating the Shapley values.
`RegressionFSII`	Estimates the FSII values using KernelSHAP.
`RegressionFBII`	Estimates the FBII values using KernelSHAP.
`KernelSHAPIQ`	The KernelSHAP-IQ regression approximator.
`InconsistentKernelSHAPIQ`	The Inconsistent KernelSHAP-IQ regression approximator.
`ProxySPEX`	ProxySPEX (SParse EXplainer) via Fourier transform sampling.
`ProxySHAP`	ProxySHAP is a proxy-based approximator that uses a regression model to approximate the value function and applies an adjustment method to better match the true value function.
`RegressionMSR`	RegressionMSR is a proxy-based approximator that uses a regression model to approximate the value function and applies the MSR adjustment method.
`SHAPIQ`	SHAP-IQ approximator for estimating Shapley interactions.
`SVARM`	The SVARM approximator for estimating the Shapley value (SV).
`SVARMIQ`	The SVARM-IQ approximator for Shapley interactions.
`kADDSHAP`	The kADD-SHAP regression approximator for estimating the kADD-SHAP values.
`SPEX`	SPEX (SParse EXplainer) via Fourier transform sampling.
`UnbiasedKernelSHAP`	The Unbiased KernelSHAP approximator for estimating the Shapley value (SV).