Abstract
Genetic Algorithms (GAs) Use Many Hyperparameters, and Tuning These Parameters Can Determine the Optimization Performance. a GA with an Augmented Initial Population Was Proposed for Decap Optimization but It Had Convergence Issues by Getting Stuck in the Local Minimum. This Work Uses a Reinforcement Learning (RL) Approach to Adaptively Tune the Hyperparameters of GA during its Operation. with This Approach, the Agent Tries to Change the Parameters So that the GA Does Not Get Stuck in the Local Minimum. the Proposed Method Combining the RL Agent and Augmented GA Showed Better Performance in Terms of Solution Quality and Time Cost. overall, in All the Cases Tested, the Proposed Method Showed Better Performance Than the Augmented GA Without RL.
Recommended Citation
H. Manoharan et al., "Augmented Genetic Algorithm V2 with Reinforcement Learning for PDN Decap Optimization," 2023 IEEE Symposium on Electromagnetic Compatibility and Signal/Power Integrity, EMC+SIPI 2023, pp. 255 - 258, Institute of Electrical and Electronics Engineers, Jan 2023.
The definitive version is available at https://doi.org/10.1109/EMCSIPI50001.2023.10241752
Department(s)
Electrical and Computer Engineering
Keywords and Phrases
Augmented Genetic Algorithm; Decap Optimization; Genetic Algorithm; Reinforcement Learning (RL)
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2023 Institute of Electrical and Electronics Engineers, All rights reserved.
Publication Date
01 Jan 2023
Comments
National Science Foundation, Grant IIP-1916535