2024年5月，国科大密码学院孙思维教授作为第三作者，与华东师范大学的王高、王高丽教授共同发表了题为“A New (Related-Key) Neural Distinguisher Using Two Differences for Differential Cryptanalysis”的研究论文，并被国际权威信息安全期刊IET Information Security接收。
Abstract: This paper presents a novel neural distinguisher for differential cryptanalysis, specifically designed for the related-key setting. Unlike traditional neural distinguishers that rely on a single difference, our approach leverages two input differences to enhance the distinguishing capability. Experimental results demonstrate that the proposed method achieves superior performance on multiple lightweight block ciphers. Compared to existing neural distinguishers, our approach more effectively captures the nonlinear characteristics of cryptographic algorithms and improves the attack success rate. This study provides new insights into deep learning-based cryptanalysis and further advances the application of artificial intelligence in the security evaluation of symmetric ciphers.
论文信息：Gao Wang, Gaoli Wang, Siwei Sun, “A New (Related-Key) Neural Distinguisher Using Two Differences for Differential Cryptanalysis,” IET Information Security (SCI，信息安全领域权威期刊).
(原文链接：https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/2024/4097586)
