AI Paper Atlas

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

NEO-KD: Knowledge-Distillation-Based Adversarial Training for Robust Multi-Exit Neural Networks

Seokil Ham, Jungwuk Park, Dong-Jun Han, Jaekyun Moon

NeurIPS 2023 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training with feature consistency · 3 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
NEO-KD instead of plain adversarial trainingMSDNet · CIFAR-10 max-averageadv. accuracy (avg): 45.91% -> 48.30%Improved
NEO-KD instead of plain adversarial trainingMSDNet · CIFAR-10 average attackadv. accuracy (avg): 40.04% -> 44.20%Improved
NEO-KD instead of plain adversarial trainingMSDNet · CIFAR-100 average attacktop-1 adv. accuracy: 18.13 -> 22.88Improved
NEO-KD instead of plain adversarial trainingMSDNet · Tiny-ImageNet average attacktop-1 adv. accuracy: 27.61 -> 30.56Improved
NEO-KD instead of plain adversarial trainingMSDNet · ImageNet average attacktop-1 adv. accuracy: 25.58 -> 29.98Improved
NEO-KD instead of plain adversarial trainingSmallCNN · MNIST max-averageadv. accuracy (avg): 94.15% -> 94.49%Improved
add self-knowledge distillation (SKD)adversarial training · CIFAR-10 max-averageadv. accuracy (avg): 45.91% -> 44.69%Hurt
add LW regularizeradversarial training · CIFAR-10 max-averageadv. accuracy (avg): 45.91% -> 35.87%Hurt
add only NKD (remove EOKD) (ablation)NEO-KD · CIFAR-10 average attackadv. accuracy (avg): 40.04% -> 40.97%Improved
add only EOKD (remove NKD) (ablation)NEO-KD · CIFAR-10 average attackadv. accuracy (avg): 40.04% -> 39.01%Hurt
ensemble neighbor exits instead of no ensembling (ablation)NKD · CIFAR-10 average attackadv. accuracy (avg): 40.49% -> 44.20%Improved
ensemble all exits instead of neighbors (ablation)NKD · CIFAR-10 average attackadv. accuracy (avg): 44.20% -> 40.06%Hurt

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