A General Taylor Framework for Unifying and Revisiting Attribution Methods

TL;DR

This paper introduces a Taylor-based theoretical framework to unify and analyze various attribution methods for deep learning models, providing insights into their principles, limitations, and performance.

Contribution

It proposes a general Taylor attribution framework that unifies 14 attribution methods and establishes principles for effective attribution, bridging theoretical understanding and practical evaluation.

Findings

Reformulates 14 attribution methods within the Taylor framework

Identifies three key principles for good attribution methods

Shows a positive correlation between principles followed and attribution performance

Abstract

Attribution methods provide an insight into the decision-making process of machine learning models, especially deep neural networks, by assigning contribution scores to each individual feature. However, the attribution problem has not been well-defined, which lacks a unified guideline to the contribution assignment process. Furthermore, existing attribution methods often built upon various empirical intuitions and heuristics. There still lacks a general theoretical framework that not only can offer a good description of the attribution problem, but also can be applied to unifying and revisiting existing attribution methods. To bridge the gap, in this paper, we propose a Taylor attribution framework, which models the attribution problem as how to decide individual payoffs in a coalition. Then, we reformulate fourteen mainstream attribution methods into the Taylor framework and analyze these attribution methods in terms of rationale, fidelity, and limitation in the framework. Moreover, we establish three principles for a good attribution in the Taylor attribution framework, i.e., low approximation error, correct Taylor contribution assignment, and unbiased baseline selection. Finally, we empirically validate the Taylor reformulations and reveal a positive correlation between the attribution performance and the number of principles followed by the attribution method via benchmarking on real-world datasets.