AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Formalizing Generalization and Adversarial Robustness of Neural Networks to Weight Perturbations

Yu-Lin Tsai, Chia-Yi Hsu, Chia-Mu Yu, Pin-Yu Chen

NeurIPS 2021 · Adversarial robustness and security › Certified and empirical robustness · approach: Robustness-accuracy trade-off analysis · 13 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
Add both the weight-perturbation robustness term and matrix-norm regularization to the classification lossMNIST feed-forward network · MNIST, weight PGD attack (200 steps; varying ε_test)robust-accuracy AUC: about 2× larger than that of the standard modelImproved
Add the weight-perturbation robustness term alone (ablation)MNIST feed-forward network · MNIST, weight PGD attack (200 steps; varying ε_test)robust accuracyImproved
Add matrix-norm generalization-gap regularization alone (ablation)MNIST feed-forward network · MNIST, weight PGD attack (200 steps; varying ε_test)robust accuracyImproved
Increase the matrix-norm regularization coefficient μ (ablation)Proposed-loss-trained MNIST network · MNIST, empirical generalization gap under weight PGD attackgeneralization gapImproved
Replace the proposed matrix-norm regularization with L1 or L2 weight decay (ablation)Proposed-loss-trained MNIST network · MNIST, empirical generalization gap and weight-perturbation robustnessgeneralization gapHurt
Use naive adversarial weight training instead of the proposed theory-driven lossMNIST feed-forward network · MNIST, weight PGD attack at different ε_testtest accuracyHurt
Train with the proposed loss rather than the standard lossMNIST network · MNIST, 2-bit weight quantizationaccuracy decrease: 3x less than the standard modelImproved

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