Beyond Single-Feature Importance with ICECREAM

TL;DR

ICECREAM introduces an information-theoretic measure to identify influential feature coalitions affecting model outputs, improving explainability and root cause analysis over existing methods.

Contribution

The paper presents ICECREAM, a novel coalition-based explanation method that captures the influence of feature groups on model outcomes, surpassing individual feature importance approaches.

Findings

ICECREAM outperforms state-of-the-art explainability methods.

It achieves high accuracy in root cause analysis.

Demonstrates effectiveness on synthetic and real-world data.

Abstract

Which set of features was responsible for a certain output of a machine learning model? Which components caused the failure of a cloud computing application? These are just two examples of questions we are addressing in this work by Identifying Coalition-based Explanations for Common and Rare Events in Any Model (ICECREAM). Specifically, we propose an information-theoretic quantitative measure for the influence of a coalition of variables on the distribution of a target variable. This allows us to identify which set of factors is essential to obtain a certain outcome, as opposed to well-established explainability and causal contribution analysis methods which can assign contributions only to individual factors and rank them by their importance. In experiments with synthetic and real-world data, we show that ICECREAM outperforms state-of-the-art methods for explainability and root cause analysis, and achieves impressive accuracy in both tasks.