ExplainReduce: Generating global explanations from many local explanations

TL;DR

ExplainReduce is a method that condenses numerous local explanations of complex models into a small, interpretable set of global explanations, making AI models more transparent and understandable.

Contribution

The paper introduces ExplainReduce, a novel optimization-based approach to generate global explanations from many local explanations, demonstrating efficiency and fidelity.

Findings

As few as five explanations can emulate complex models.

ExplainReduce is competitive with existing model aggregation methods.

The reduction process is formulated as an efficient greedy heuristic.

Abstract

Most commonly used non-linear machine learning methods are closed-box models, uninterpretable to humans. The field of explainable artificial intelligence (XAI) aims to develop tools to examine the inner workings of these closed boxes. An often-used model-agnostic approach to XAI involves using simple models as local approximations to produce so-called local explanations; examples of this approach include LIME, SHAP, and SLISEMAP. This paper shows how a large set of local explanations can be reduced to a small "proxy set" of simple models, which can act as a generative global explanation. This reduction procedure, ExplainReduce, can be formulated as an optimisation problem and approximated efficiently using greedy heuristics. We show that, for many problems, as few as five explanations can faithfully emulate the closed-box model and that our reduction procedure is competitive with other model aggregation methods.