Year: 2023
Type: conference-paper
Abstract: Several recent methods for interpretability model feature interactions by looking at the Hessian of a neural network. This poses a challenge for ReLU networks, which are piecewise-linear and thus have a zero Hessian almost everywhere. We propose SmoothHess, a method of estimating second-order interactions through Stein's Lemma. In particular, we estimate the Hessian of the network convolved with a Gaussian through an efficient sampling algorithm, requiring only network gradient calls. SmoothHess is applied post-hoc, requires no modifications to the ReLU network architecture, and the extent of smoothing can be controlled explicitly. We provide a non-asymptotic bound on the sample complexity of our estimation procedure. We validate the superior ability of SmoothHess to capture interactions on benchmark datasets and a real-world medical spirometry dataset.
DOI: 10.52202/075280-2206
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: Confocal video-mosaicking microscopy to guide surgery of superficially spreading skin cancers, G289723095, Cancer Center Support Grant, SYSTEMS APPROACHES TO THE EPIDEMIOLOGY, GENETICS AND GENOMICS OF LUNG DISEASES, M-ISIC: A Multimodal Open-Source International Skin Imaging Collaboration Informatics Platform for Automated Skin Cancer Detection
OpenAlex ID: W7133215570