Author(s): Yifei Zhang, Dong Chen, Fan Wang, Wenrui Zhang, Yan Chen, Dingding Han, Jianmin Yuan, Xiangjin Kong, and Yu-Gang Ma We demonstrate that AI-driven program synthesis can autonomously discover fundamental strategies for decomposing unitary matrices in photonic networks. By extending DreamCoder to complex-valued linear algebra, the system generates decomposition programs achieving the minimal $N(N−1)/2$ Mach-Zehnder … [Phys. Rev. Lett. 137, 043801] Published Mon Jul 20, 2026