Looking From the Future: Multi-order Iterations Can Enhance Adversarial Attack Transferability

TL;DR

This paper introduces a novel sequence optimization approach called Looking From the Future (LFF) to improve the transferability of adversarial attacks, demonstrating significant enhancements on ImageNet1k with state-of-the-art methods.

Contribution

The paper proposes the LFF concept and extends it to multi-order attacks, significantly improving adversarial transferability compared to existing methods.

Findings

LFF-based attacks outperform baseline methods in transferability.

Multi-order LFF attacks further enhance attack success rates.

Experimental results confirm the effectiveness of the proposed approach.

Abstract

Various methods try to enhance adversarial transferability by improving the generalization from different perspectives. In this paper, we rethink the optimization process and propose a novel sequence optimization concept, which is named Looking From the Future (LFF). LFF makes use of the original optimization process to refine the very first local optimization choice. Adapting the LFF concept to the adversarial attack task, we further propose an LFF attack as well as an MLFF attack with better generalization ability. Furthermore, guiding with the LFF concept, we propose an $LLF^{\mathcal{N}}$ attack which entends the LFF attack to a multi-order attack, further enhancing the transfer attack ability. All our proposed methods can be directly applied to the iteration-based attack methods. We evaluate our proposed method on the ImageNet1k dataset by applying several SOTA adversarial attack methods under four kinds of tasks. Experimental results show that our proposed method can greatly enhance the attack transferability. Ablation experiments are also applied to verify the effectiveness of each component. The source code will be released after this paper is accepted.