Clean Test Accuracy and Adversarial Training via Average Attack

by Writings, Papers and Blogs on Text ModelsSeptember 30th, 2024

Too Long; Didn't Read

The Clean Test Accuracy section reveals that NEO-KD maintains comparable clean accuracy to Adv. w/o Distill, particularly on the Tiny-ImageNet dataset, while significantly improving adversarial test accuracy. The Adversarial Training via Average Attack section discusses additional experiments using average attacks, confirming NEO-KD’s advantages in adversarial training across various datasets, including CIFAR-100.

featured image - Clean Test Accuracy and Adversarial Training via Average Attack

Authors:

(1) Seokil Ham, KAIST;

(2) Jungwuk Park, KAIST;

(3) Dong-Jun Han, Purdue University;

(4) Jaekyun Moon, KAIST.

Table of Links

Abstract and 1. Introduction

2. Related Works

3. Proposed NEO-KD Algorithm and 3.1 Problem Setup: Adversarial Training in Multi-Exit Networks

3.2 Algorithm Description

4. Experiments and 4.1 Experimental Setup

4.2. Main Experimental Results

4.3. Ablation Studies and Discussions

5. Conclusion, Acknowledgement and References

A. Experiment Details

B. Clean Test Accuracy and C. Adversarial Training via Average Attack

D. Hyperparameter Tuning

E. Discussions on Performance Degradation at Later Exits

F. Comparison with Recent Defense Methods for Single-Exit Networks

G. Comparison with SKD and ARD and H. Implementations of Stronger Attacker Algorithms

B Clean Test Accuracy

Table A1 shows the clean accuracy results using the model built upon adversarial training via maxaverage attack. We observe that NEO-KD generally shows comparable clean test accuracy with Adv. w/o Distill [12], especially on the more complicated dataset Tiny-ImageNet [17] while achieving much better adversarial test accuracy as reported in the main manuscript.

C Adversarial Training via Average Attack

In the main manuscript, we presented experimental results using the model trained based on maxaverage attack. Here, we also adversarially train the model via average attack [12] and measure adversarial test accuracy on CIFAR-100 dataset. Table A2 compares adversarial test accuracies of NEO-KD and other baselines against max-average attack and average attack. The overall results are consistent with the ones in the main manuscript with adversarial training via max-average attack, further confirming the advantage of NEO-KD.