Assessment of Reinforcement Learning for Macro Placement
Assessment of Reinforcement Learning for Macro Placement
复制标题
强化学习对宏观布局的评估
DOI：
10.1145/3569052.3578926
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Wang, Zhiang
中科院分区：
文献类型：
--
作者：
Cheng, Chung-Kuan;Kahng, Andrew B.;Kundu, Sayak;Wang, Yucheng;Wang, Zhiang
我们在GitHub中提供开放、透明的实施和评估Google Brain的深度强化学习方法，用于宏放置（Nature）及其电路训练（CT）实施。我们在开源中实现CT的关键“黑盒”元素，并澄清CT和自然之间的差异。开发并发布了关于开放启用的新测试用例。我们与多个替代宏放置器一起评估CT，所有评估流程和相关脚本都在GitHub中公开。我们的实验还包括学术混合尺寸放置基准，以及消融和稳定性研究。我们评论自然和CT的影响，以及未来的研究方向。
We provide open, transparent implementation and assessment of Google Brain's deep reinforcement learning approach to macro placement (Nature) and its Circuit Training (CT) implementation in GitHub. We implement in open-source key "blackbox" elements of CT, and clarify discrepancies between CT and Nature. New testcases on open enablements are developed and released. We assess CT alongside multiple alternative macro placers, with all evaluation flows and related scripts public in GitHub. Our experiments also encompass academic mixed-size placement benchmarks, as well as ablation and stability studies. We comment on the impact of Nature and CT, as well as directions for future research.
RosettaStone：连接物理设计研究的过去、现在和未来
影响因子：
2
作者：
A. Kahng;Minsoo Kim;Seungwon Kim;M. Woo
通讯作者：
M. Woo