arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.08657cs.NE
收藏

NEVO-GSPT:基于 Inflate 和 Deflate 操作符的群体神经网络进化

NEVO-GSPT: Population-Based Neural Network Evolution Using Inflate and Deflate Operators

NOVA Information Management School (NOVA IMS), Universidade Nova de Lisboa(里斯本新大学信息管理学院);Vita-Salute San Raffaele University, Milan, Italy(米兰 Vita-Salute 圣拉斐尔大学);Department of Engineering and Architecture, University of Trieste, Italy(的里雅斯特大学工程与建筑系);Comprehensive Cancer Center/Unit of Urology(米兰圣拉斐尔医院综合癌症中心/泌尿科);URI;IRCCS Ospedale San Raffaele, Milan, Italy

Davide Farinati, Frederico J. J. B. Santos, Leonardo Vanneschi, Mauro Castelli

AI总结:

NEVO-GSPT通过引入 Inflate和Deflate操作符,实现高效神经网络进化,生成性能优异的紧凑型网络。

摘要:

进化神经网络架构是一个计算需求很高的过程。传统方法往往需要在大型架构空间中进行广泛的搜索,并且对结构修改如何影响模型行为的理解有限。本文介绍了一个新的神经进化算法 NGSP T,基于两个关键创新。首先,我们从遗传编程中适应了几何语义操作符(GSOs)到神经网络进化,确保架构变化在单峰误差曲面上产生可预测的效果。其次,我们引入了一个新的操作符(DGSM),使网络大小能够受控减少,同时保持 GSOs 的语义属性。与传统方法不同,NGSP T 的高效评估机制仅需计算新添加组件的语义,允许高效的群体训练,从而在计算成本仅为传统方法一小部分的情况下,对搜索空间进行全面探索。在四个回归基准测试中,实验结果表明,NGSP T 一致地进化出紧凑的神经网络,其性能与文献中已建立的方法如标准神经网络、SLIM-GSGP、TensorNEAT 和 SLM 相当或更好。

英文摘要:

Evolving neural network architectures is a computationally demanding process. Traditional methods often require an extensive search through large architectural spaces and offer limited understanding of how structural modifications influence model behavior. This paper introduces \gls{ngspt}, a novel Neuroevolution algorithm based on two key innovations. First, we adapt geometric semantic operators~(GSOs) from genetic programming to neural network evolution, ensuring that architectural changes produce predictable effects on network semantics within a unimodal error surface. Second, we introduce a novel operator (DGSM) that enables controlled reduction of network size, while maintaining the semantic properties of~GSOs. Unlike traditional approaches, \gls{ngspt}'s efficient evaluation mechanism, which only requires computing the semantics of newly added components, allows for efficient population-based training, resulting in a comprehensive exploration of the search space at a fraction of the computational cost. Experimental results on four regression benchmarks show that \gls{ngspt} consistently evolves compact neural networks that achieve performance comparable to or better than established methods in the literature, such as standard neural networks, SLIM-GSGP, TensorNEAT, and SLM.

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。

↑