Adversarial Ensemble Training by Jointly Learning Label Dependencies and Member Models
Not on the map. Why?
Abstract (arXiv)
Training an ensemble of diverse sub-models has been empirically demonstrated as an effective strategy for improving the adversarial robustness of deep neural networks. However, current ensemble training methods for image recognition typically encode image labels using one-hot vectors, which overlook dependency relationships between the labels. In this paper, we propose a novel adversarial en-semble training approach that jointly learns the label dependencies and member models. Our approach adaptively exploits the learned label dependencies to pro-mote diversity among the member models. We evaluate our approach on widely used datasets including MNIST, FashionMNIST, and CIFAR-10, and show that it achieves superior robustness against black-box attacks compared to state-of-the-art methods. Our code is available at https://github.com/ZJLAB-AMMI/LSD.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 35 of its 46 references that are in the map. First posted on arXiv 29 Jun 2022.
Builds on
35 of 46 references.
References · 35
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
- 2014Explaining and Harnessing Adversarial Examplescited by 7,364
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
- 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
- 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
- 2016Adversarial examples in the physical worldcited by 1,769
- 2016End to End Learning for Self-Driving Carscited by 1,009
- 2019When Does Label Smoothing Help?cited by 988
23 more
- 2019Adversarial Examples Are Not Bugs, They Are Featurescited by 953
- 2015The Limitations of Deep Learning in Adversarial Settingscited by 790
- 2016Delving into Transferable Adversarial Examples and Black-box Attackscited by 536
- 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 446
- 2014Deep Speech: Scaling up end-to-end speech recognitioncited by 354
- 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 351
- 2017Recent Trends in Deep Learning Based Natural Language Processingcited by 263
- 2018Generating Adversarial Examples with Adversarial Networkscited by 258
- 2019Adversarial Attacks and Defenses in Images, Graphs and Text: A Reviewcited by 237
- 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
- 2018Event-based Vision meets Deep Learning on Steering Prediction for Self-driving Carscited by 200
- 2018Spatially Transformed Adversarial Examplescited by 195
- 2015Label-Embedding for Image Classificationcited by 165
- 2020DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensemblescited by 73
- 2019Improving Adversarial Robustness of Ensembles with Diversity Trainingcited by 69
- 2019advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorchcited by 69
- 2020Transferable Perturbations of Deep Feature Distributionscited by 51
- 2021TRS: Transferability Reduced Ensemble via Encouraging Gradient Diversity and Model Smoothnesscited by 38
- 2020EMPIR: Ensembles of Mixed Precision Deep Networks for Increased Robustness against Adversarial Attackscited by 28
- 2021Diversifying Dialog Generation via Adaptive Label Smoothingcited by 25
- 2020Label Confusion Learning to Enhance Text Classification Modelscited by 16
- 2021Enhancing Label Correlation Feedback in Multi-Label Text Classification via Multi-Task Learningcited by 15
- 2020Label Smoothing and Adversarial Robustnesscited by 10
Built on by
Citing papers per year: 2023: 1; 2024: 0; 2025: 1.
2 papers.