AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

A Unified Taylor Framework for Revisiting Attribution Methods

Huiqi Deng, Na Zou, Mengnan Du, Weifu Chen, Guocan Feng, Xia Hu

AAAI 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Provable post-hoc explanations · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Include second-order Taylor termsTaylor explanation · MNIST, C-sginfidelity: 0.48 -> 0.27Improved
Include second-order Taylor termsTaylor explanation · MNIST, C-spinfidelity: 0.84 -> 0.75Improved
Use Expected Gradients with multiple Gaussian-noise baselines instead of a single fixed baselineIntegrated Gradients · ImageNet object localizationbounding box accuracyImproved
Follow more of the three attribution principlesAttribution methods · MNISTinfidelityImproved
Follow more of the three attribution principlesAttribution methods · ImageNetbounding box accuracyImproved
Use a black image as the baselineIntegrated Gradients · ImageNet corgi visualizationfeature attribution fidelityHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy