Reinforcement Learning for Efficient Design Space Exploration With Variable Fidelity Analysis Models

Author:

Agrawal Akash1,McComb Christopher1

Affiliation:

1. Carnegie Mellon University Department of Mechanical Engineering, , 5000 Forbes Avenue, Pittsburgh, PA 15213

Abstract

Abstract Reinforcement learning algorithms can autonomously learn to search a design space for high-performance solutions. However, modern engineering often entails the use of computationally intensive simulation, which can lead to slower design timelines with highly iterative approaches such as reinforcement learning. This work provides a reinforcement learning framework that leverages models of varying fidelity to enable an effective solution search while reducing overall computational needs. Specifically, it utilizes models of varying fidelity while training the agent, iteratively progressing from low- to high fidelity. To demonstrate the effectiveness of the proposed framework, we apply it to two multimodal multi-objective constrained mixed integer nonlinear design problems involving the components of a ground and aerial vehicle. Specifically, for each problem, we utilize a high-fidelity and a low-fidelity deep neural network surrogate model, trained on performance data generated from underlying ground truth models. A tradeoff between solution quality and the proportion of low-fidelity surrogate model usage is observed. Specifically, high-quality solutions are achieved with substantial reductions in computational expense, showcasing the effectiveness of the framework for design problems where the use of just a high-fidelity model is infeasible. This solution quality-computational efficiency tradeoff is contextualized by visualizing the exploration behavior of the design agents.

Funder

Defense Advanced Research Projects Agency

Publisher

ASME International

Subject

Industrial and Manufacturing Engineering,Computer Graphics and Computer-Aided Design,Computer Science Applications,Software

Reference49 articles.

1. Learning to Optimize;Li,2016

2. A Case Study of Deep Reinforcement Learning for Engineering Design: Application to Microfluidic Devices for Flow Sculpting;Lee;ASME J. Mech. Des.,2019

3. Reinforcement Learning for Engineering Design Automation;Dworschak;Adv. Eng. Inform.,2022

4. Design Synthesis Through a Markov Decision Process and Reinforcement Learning Framework;Ororbia;ASME J. Comput. Inf. Sci. Eng.,2022

5. On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?;Bender,2021

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. A Cost-Aware Multi-Agent System for Black-Box Design Space Exploration;Journal of Mechanical Design;2024-08-21

2. Leveraging deep reinforcement learning for design space exploration with multi-fidelity surrogate model;Journal of Engineering Design;2024-06-27

3. Fairness- and Uncertainty-Aware Data Generation for Data-Driven Design Based on Active Learning;Journal of Computing and Information Science in Engineering;2024-01-29

4. Deep Learning in Computational Design Synthesis: A Comprehensive Review;Journal of Computing and Information Science in Engineering;2024-01-08

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3