Attacking Compressed Vision Transformers

Swapnil Parekh, Pratyush Shukla, Devansh Shah

Abstract: Vision Transformers are increasingly embedded in industrial systems due to their superior performance, but their memory and power requirements make deploying them to edge devices a challenging task. Hence, model compression techniques are now widely used to deploy models on edge devices as they decrease the resource requirements and make model inference very fast and efficient. But their reliability and robustness from a security perspective are major issues in safety-critical applications. Adversarial attacks are like optical illusions for ML algorithms and they can severely impact the accuracy and reliability of models. In this work, we investigate the performance of adversarial attacks across the Vision Transformer model compressed using 3 SOTA compression techniques. We also analyze the effect different compression techniques like Quantization, Pruning, and Weight Multiplexing have on the transferability of adversarial attacks.

External IDs:dblp:conf/ficc/ParekhSS23