Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings
From MaRDI portal
Publication:6380446
arXiv2110.08105MaRDI QIDQ6380446
Author name not available (Why is that?)
Publication date: 15 October 2021
Abstract: We study the effects of constrained optimization formulations and Frank-Wolfe algorithms for obtaining interpretable neural network predictions. Reformulating the Rate-Distortion Explanations (RDE) method for relevance attribution as a constrained optimization problem provides precise control over the sparsity of relevance maps. This enables a novel multi-rate as well as a relevance-ordering variant of RDE that both empirically outperform standard RDE and other baseline methods in a well-established comparison test. We showcase several deterministic and stochastic variants of the Frank-Wolfe algorithm and their effectiveness for RDE.
Has companion code repository: https://github.com/zib-iol/fw-rde
This page was built for publication: Interpretable Neural Networks with Frank-Wolfe: Sparse Relevance Maps and Relevance Orderings
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6380446)
