arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.01313cs.ETcs.NEphysics.optics
收藏

ADEPT-Z:面向帕累托最优光子张量核的零样本自动化电路拓扑搜索

ADEPT-Z: Zero-Shot Automated Circuit Topology Search for Pareto-Optimal Photonic Tensor Cores

Ziyang Jiang, Pingchuan Ma, Meng Zhang, Rena Huang, Jiaqi Gu

AI总结:

针对光子张量核人工设计及梯度拓扑搜索效率低、多目标优化难的问题,本文提出零样本多目标进化框架 ADEPT-Z,在复杂硬件约束下快速搜索帕累托最优设计。

摘要:

光子张量核(PTC)是基于可编程光子集成电路的光学人工智能(AI)加速器的关键构建模块。如今,大多数 PTC 设计都由人工构建,设计效率较低,且解决方案质量不能令人满意。这使得其难以满足各种硬件规格,也难以跟上快速演进的 AI 应用。先前工作已经探索了基于梯度的方法,以可微方式学习良好的 PTC 结构。然而,在处理多个不可微目标和约束时,该方法存在训练速度慢和优化困难的问题。因此,本文提出更灵活、高效的零样本多目标进化拓扑搜索框架 ADEPT-Z,在更大的搜索空间中利用先进器件探索帕累托最优 PTC 设计。该框架能够在遵守复杂硬件约束的同时协同优化多个目标。仅需不到 3 小时的搜索,即可获得数十个多样化的帕累托最优解决方案,比先前基于梯度的方法快 100 倍;其精度加权面积-能效比先前人工设计高 2 倍,性能优于这些设计。ADEPT-Z 的代码可在 https://github.com/ScopeX-ASU/ADEPT-Z 获取。

英文摘要:

Photonic tensor cores (PTCs) are essential building blocks for optical artificial intelligence (AI) accelerators based on programmable photonic integrated circuits. Most PTC designs today are manually constructed, with low design efficiency and unsatisfying solution quality. This makes it challenging to meet various hardware specifications and keep up with rapidly evolving AI applications. Prior work has explored gradient-based methods to learn a good PTC structure differentiably. However, it suffers from slow training speed and optimization difficulty when handling multiple non-differentiable objectives and constraints. Therefore, in this work, we propose a more flexible and efficient zero-shot multi-objective evolutionary topology search framework ADEPT-Z that explores Pareto-optimal PTC designs with advanced devices in a larger search space. Multiple objectives can be co-optimized while honoring complicated hardware constraints. With only <3 hours of search, we can obtain tens of diverse Pareto-optimal solutions, 100x faster than the prior gradient-based method, outperforming prior manual designs with 2x higher accuracy weighted area-energy efficiency. The code of ADEPT-Z is available at https://github.com/ScopeX-ASU/ADEPT-Z.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑