A logic optimization command sequence combination method based on game reinforcement learning

By introducing a game-theoretic reinforcement learning-based logic optimization method, and utilizing two agents and Shapley value allocation, the contradiction between area and delay in integrated circuit logic optimization is resolved, achieving the optimal trade-off in logic optimization and improving design efficiency and accuracy.

CN119514440BActive Publication Date: 2026-02-06JIANGNAN UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202411328678.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-24
Publication Date
2026-02-06
Estimated Expiration
2044-09-24

AI Technical Summary

Technical Problem

Existing technologies struggle to achieve an optimal trade-off between area and delay in integrated circuit logic optimization, leading to contradictions and inconsistencies during the optimization process.

Method used

Two agents are introduced, one for optimizing circuit delay and the other for optimizing area. A game-theoretic reinforcement learning architecture is constructed using the A2C algorithm. The optimal combination of command sequences for logic optimization is achieved through a distributed action network and a centralized evaluation network, combined with Shapley value allocation.

Benefits of technology

It effectively resolves the conflict between area and delay, finds the optimal trade-off solution for logic optimization, and improves the efficiency and accuracy of logic circuit design and optimization.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a logic optimization command sequence combination method based on game reinforcement learning and belongs to the technical field of integrated circuit design automation. Two agents are introduced, and a game reinforcement learning framework is constructed by adopting an A2C algorithm between the two agents. A first agent is trained by taking a delay optimization target reward and a global optimization target sharpe value distribution as a reward function, and a second agent is trained by taking an area optimization target reward and a global optimization target sharpe value distribution as a reward function. A decentralized action network allows the agents to have different optimization tendencies, and a centralized evaluation network allows the agents to focus on the global situation and coordinate optimization towards the overall target. In addition, the application introduces a sharpe value to realize fair distribution of the overall reward among the agents, and solves the problem that an optimal trade-off solution of the area and delay optimization targets cannot be found in the prior art.
Need to check novelty before this filing date? Find Prior Art