MARGIN: Uncovering Deep Neural Networks using Graph Signal Analysis
This work addresses the need for trust and insights in complex neural networks for users, though it appears incremental as it builds on existing graph analysis techniques.
The paper tackles the problem of interpretability in deep neural networks by introducing MARGIN, a general approach based on graph signal analysis that identifies influential nodes to address various interpretability tasks, and it demonstrates superior performance over existing methods across multiple challenges.
Interpretability has emerged as a crucial aspect of building trust in machine learning systems, aimed at providing insights into the working of complex neural networks that are otherwise opaque to a user. There are a plethora of existing solutions addressing various aspects of interpretability ranging from identifying prototypical samples in a dataset to explaining image predictions or explaining mis-classifications. While all of these diverse techniques address seemingly different aspects of interpretability, we hypothesize that a large family of interepretability tasks are variants of the same central problem which is identifying \emph{relative} change in a model's prediction. This paper introduces MARGIN, a simple yet general approach to address a large set of interpretability tasks MARGIN exploits ideas rooted in graph signal analysis to determine influential nodes in a graph, which are defined as those nodes that maximally describe a function defined on the graph. By carefully defining task-specific graphs and functions, we demonstrate that MARGIN outperforms existing approaches in a number of disparate interpretability challenges.