Work

MapTune: Advancing ASIC Technology Mapping via Reinforcement Learning Guided Library Tuning

Year: 2024

Type: conference-paper

Abstract: Technology mapping involves mapping logical circuits to a library of cells. Traditionally, the full technology library is used, leading to a large search space and potential overhead. Motivated by randomly sampled technology mapping case studies, we propose MapTune framework that addresses this challenge by utilizing reinforcement learning to make design-specific choices during cell selection. By learning from the environment, MapTune refines the cell selection process, resulting in a reduced search space and potentially improved mapping quality.

DOI: 10.1145/3676536.3676762

Authors: Mingju Liu, Daniel Robinson, Yingjie Li, Cunxi Yu

Institutions: University of Maryland, College Park, Massachusetts Institute of Technology

Language: English

Topic: VLSI and FPGA Design Techniques

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Funder: National Science Foundation

Awards: Collaborative Research: SWIFT: Decentralized Intelligent Spectrum Sharing in UAV Networks (DISH-uNET) via Hardware-software Co-design, CAREER: OneSense: One-Rule-for-All Combinatorial Boolean Synthesis via Reinforcement Learning, Collaborative Research: FMitF: Track I: DeepSmith: Scheduling with Quality Guarantees for Efficient DNN Model Execution, Collaborative Research: SHF: Medium: Differentiable Hardware Synthesis