Multi-agent Reinforcement Learning with Auto Group Assigning for Practical Analog-LDO Sizing
Published in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), 2025
This paper presents a multi-agent reinforcement learning framework with automatic group assignment for practical analog-LDO sizing. It introduces a novel device grouping method based on current transient characteristics, achieving 5.2× faster training convergence compared to conventional module-based partitioning approaches.
Recommended citation: Wu, H., Jiang, H., Wang, Z., Ou, Y., Yuan, B., Lu, Y., & Jiang, J. (2025). "Multi-agent Reinforcement Learning with Auto Group Assigning for Practical Analog-LDO Sizing." IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD). (Under review). #