Adaptive Pruning with Module Robustness Sensitivity: Balancing Compression and Robustness
This work addresses the trade-off between model compression and robustness preservation for neural network practitioners, offering a practical and generalizable framework.
The paper tackled the problem of neural network pruning often compromising adversarial robustness, and introduced Module Robustness Sensitivity (MRS) to balance compression and robustness, resulting in significant robustness enhancements while maintaining competitive accuracy and efficiency across multiple datasets and architectures.
Neural network pruning has traditionally focused on weight-based criteria to achieve model compression, frequently overlooking the crucial balance between adversarial robustness and accuracy. Existing approaches often fail to preserve robustness in pruned networks, leaving them more susceptible to adversarial attacks. This paper introduces Module Robustness Sensitivity (MRS), a novel metric that quantifies layer-wise sensitivity to adversarial perturbations and dynamically informs pruning decisions. Leveraging MRS, we propose Module Robust Pruning and Fine-Tuning (MRPF), an adaptive pruning algorithm compatible with any adversarial training method, offering both flexibility and scalability. Extensive experiments on SVHN, CIFAR, and Tiny-ImageNet across diverse architectures, including ResNet, VGG, and MobileViT, demonstrate that MRPF significantly enhances adversarial robustness while maintaining competitive accuracy and computational efficiency. Furthermore, MRPF consistently outperforms state-of-the-art structured pruning methods in balancing robustness, accuracy, and compression. This work establishes a practical and generalizable framework for robust pruning, addressing the long-standing trade-off between model compression and robustness preservation.