Trust-Region Method with Deep Reinforcement Learning in Analog Design Space Exploration

This paper introduces new perspectives on analog design space search. To minimize the time-to-market, this endeavor better cast as constraint satisfaction problem than global optimization defined in prior arts. We incorporate model based agents, contrasted with model-free learning, to implement a tr...

Full description

Saved in:

Bibliographic Details
Published in:	2021 58th ACM/IEEE Design Automation Conference (DAC) pp. 1225 - 1230
Main Authors:	Yang, Kai-En, Tsai, Chia-Yu, Shen, Hung-Hao, Chiang, Chen-Feng, Tsai, Feng-Ming, Wang, Chung-An, Ting, Yiju, Yeh, Chia-Shun, Lai, Chin-Tang
Format:	Conference Proceeding
Language:	English
Published:	IEEE 05.12.2021
Subjects:	artificial intelligence Design automation electronic design automation Employee welfare Production Reinforcement learning Search problems Space exploration Supervised learning transistor sizing
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	This paper introduces new perspectives on analog design space search. To minimize the time-to-market, this endeavor better cast as constraint satisfaction problem than global optimization defined in prior arts. We incorporate model based agents, contrasted with model-free learning, to implement a trust-region strategy. As such, simple feed-forward networks can be trained with supervised learning, where the convergence is relatively trivial. Experiment results demonstrate orders of magnitude improvement on search iterations. Additionally, the unprecedented consideration of PVT conditions are accommodated. On circuits with TSMC 5/6nm process, our method achieve performance surpassing human designers. Furthermore, this framework is in production in industrial settings.
DOI:	10.1109/DAC18074.2021.9586087