arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.01145cs.NEcs.AI
收藏

ReEvo:作为具有反思进化能力的超启发式的大型语言模型

ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution

Peking University(北京大学);Southeast University(东南大学);Singapore Management University(新加坡管理大学);KAIST(韩国科学技术院);PKU-Wuhan Institute for Artificial Intelligence(北京大学武汉人工智能研究院)

Haoran Ye, Jiarui Wang, Zhiguang Cao, Federico Berto, Chuanbo Hua, Haeyeon Kim, Jinkyoo Park, Guojie Song

AI总结:

本文提出Reflective Evolution(ReEvo),通过结合进化搜索与LLM反思来赋能Language Hyper-Heuristics,在多种组合优化问题中高效生成最先进的启发式算法,且比以往方法更具样本效率。

摘要:

NP-hard组合优化问题(COPs)的普遍存在迫使领域专家进行试错式的启发式设计。随着大型语言模型(LLMs)的兴起,设计自动化的长期努力获得了新动力。本文引入了Language Hyper-Heuristics(LHHs),这是一种超启发式的新兴变体,利用LLMs进行启发式生成,具有最少的人工干预和开放式启发式空间。为了赋能LHHs,我们提出了Reflective Evolution(ReEvo),这是一种新颖的集成方法,结合了用于高效探索启发式空间的进化搜索,以及在该空间内提供语言梯度的LLM反思。在五种异构算法类型、六种不同的COPs,以及COPs的白盒和黑盒视角下,ReEvo产生了最先进且具竞争力的元启发式算法、进化算法、启发式算法和神经求解器,同时比先前的LHHs具有更高的样本效率。

英文摘要:

The omnipresence of NP-hard combinatorial optimization problems (COPs) compels domain experts to engage in trial-and-error heuristic design. The long-standing endeavor of design automation has gained new momentum with the rise of large language models (LLMs). This paper introduces Language Hyper-Heuristics (LHHs), an emerging variant of Hyper-Heuristics that leverages LLMs for heuristic generation, featuring minimal manual intervention and open-ended heuristic spaces. To empower LHHs, we present Reflective Evolution (ReEvo), a novel integration of evolutionary search for efficiently exploring the heuristic space, and LLM reflections to provide verbal gradients within the space. Across five heterogeneous algorithmic types, six different COPs, and both white-box and black-box views of COPs, ReEvo yields state-of-the-art and competitive meta-heuristics, evolutionary algorithms, heuristics, and neural solvers, while being more sample-efficient than prior LHHs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑