Two is Better than One: Efficient Ensemble Defense for Robust and Compact Models
Not on the map. Why?
Abstract (arXiv)
Deep learning-based computer vision systems adopt complex and large architectures to improve performance, yet they face challenges in deployment on resource-constrained mobile and edge devices. To address this issue, model compression techniques such as pruning, quantization, and matrix factorization have been proposed; however, these compressed models are often highly vulnerable to adversarial attacks. We introduce the Efficient Ensemble Defense (EED) technique, which diversifies the compression of a single base model based on different pruning importance scores and enhances ensemble diversity to achieve high adversarial robustness and resource efficiency. EED dynamically determines the number of necessary sub-models during the inference stage, minimizing unnecessary computations while maintaining high robustness. On the CIFAR-10 and SVHN datasets, EED demonstrated state-of-the-art robustness performance compared to existing adversarial pruning techniques, along with an inference speed improvement of up to 1.86 times. This proves that EED is a powerful defense solution in resource-constrained environments.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 43 of its 53 references that are in the map. First posted on arXiv 7 Apr 2025.
Builds on
43 of 53 references.
References · 43
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,291
- 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
- 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
- 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
- 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
- 2016Pruning Filters for Efficient ConvNetscited by 1,129
- 2014Two-Stream Convolutional Networks for Action Recognition in Videoscited by 1,067
31 more
- 2016Adversarial Machine Learning at Scalecited by 934
- 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
- 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
- 2022Diffusion Models for Adversarial Purificationcited by 469
- 2017NISP: Pruning Networks using Neuron Importance Score Propagationcited by 228
- 2013Predicting Parameters in Deep Learningcited by 223
- 2019Improving Adversarial Robustness via Promoting Ensemble Diversitycited by 213
- 2018Stochastic Activation Pruning for Robust Adversarial Defensecited by 171
- 2015Compressing Neural Networks with the Hashing Trickcited by 150
- 2021Adversarial purification with Score-based generative modelscited by 132
- 2019Adversarial Robustness vs Model Compression, or Both?cited by 98
- 2020Do Wider Neural Networks Really Help Adversarial Robustness?cited by 79
- 2020DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensemblescited by 73
- 2018Sparse DNNs with Improved Adversarial Robustnesscited by 71
- 2019Improving Adversarial Robustness of Ensembles with Diversity Trainingcited by 69
- 2019Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-High Resolution Imagescited by 51
- 2022Sparsity Winning Twice: Better Robust Generalization from More Efficient Trainingcited by 41
- 2020Wisdom of Committees: An Overlooked Approach To Faster and More Accurate Modelscited by 39
- 2020A Tunable Robust Pruning Framework Through Dynamic Network Rewiring of DNNscited by 37
- 2019Dual Dynamic Inference: Enabling More Efficient, Adaptive and Controllable Deep Inferencecited by 31
- 2021Drawing Robust Scratch Tickets: Subnetworks with Inborn Robustness Are Found within Randomly Initialized Networkscited by 29
- 2022Masking Adversarial Damage: Finding Adversarial Saliency for Robust and Sparse Networkcited by 21
- 2018To compress or not to compress: Understanding the Interactions between Adversarial Attacks and Neural Network Compressioncited by 14
- 2020Achieving Adversarial Robustness via Sparsitycited by 14
- 2023Understanding and Improving Ensemble Adversarial Defensecited by 14
- 2022Pruning Adversarially Robust Neural Networks without Adversarial Examplescited by 12
- 2023Towards Inference Efficient Deep Ensemble Learningcited by 11
- 2024Adversarial Pruning: A Survey and Benchmark of Pruning Methods for Adversarial Robustnesscited by 10
- 2020Towards Practical Lottery Ticket Hypothesis for Adversarial Trainingcited by 9
- 2022Understanding the effect of sparsity on neural networks robustnesscited by 8
- 2024Holistic Adversarially Robust Pruningcited by 5
Built on by
Citing papers per year: 2025: 1.
1 paper.