神经架构搜索基准设计得好吗?对操作重要性的深入探讨

Vasco Lopes, Bruno Degardin, L. Alexandre
{"title":"神经架构搜索基准设计得好吗?对操作重要性的深入探讨","authors":"Vasco Lopes, Bruno Degardin, L. Alexandre","doi":"10.48550/arXiv.2303.16938","DOIUrl":null,"url":null,"abstract":"Neural Architecture Search (NAS) benchmarks significantly improved the capability of developing and comparing NAS methods while at the same time drastically reduced the computational overhead by providing meta-information about thousands of trained neural networks. However, tabular benchmarks have several drawbacks that can hinder fair comparisons and provide unreliable results. These usually focus on providing a small pool of operations in heavily constrained search spaces -- usually cell-based neural networks with pre-defined outer-skeletons. In this work, we conducted an empirical analysis of the widely used NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks in terms of their generability and how different operations influence the performance of the generated architectures. We found that only a subset of the operation pool is required to generate architectures close to the upper-bound of the performance range. Also, the performance distribution is negatively skewed, having a higher density of architectures in the upper-bound range. We consistently found convolution layers to have the highest impact on the architecture's performance, and that specific combination of operations favors top-scoring architectures. These findings shed insights on the correct evaluation and comparison of NAS methods using NAS benchmarks, showing that directly searching on NAS-Bench-201, ImageNet16-120 and TransNAS-Bench-101 produces more reliable results than searching only on CIFAR-10. Furthermore, with this work we provide suggestions for future benchmark evaluations and design. The code used to conduct the evaluations is available at https://github.com/VascoLopes/NAS-Benchmark-Evaluation.","PeriodicalId":13641,"journal":{"name":"Inf. Sci.","volume":"5 1","pages":"119695"},"PeriodicalIF":0.0000,"publicationDate":"2023-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Are Neural Architecture Search Benchmarks Well Designed? A Deeper Look Into Operation Importance\",\"authors\":\"Vasco Lopes, Bruno Degardin, L. Alexandre\",\"doi\":\"10.48550/arXiv.2303.16938\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Neural Architecture Search (NAS) benchmarks significantly improved the capability of developing and comparing NAS methods while at the same time drastically reduced the computational overhead by providing meta-information about thousands of trained neural networks. However, tabular benchmarks have several drawbacks that can hinder fair comparisons and provide unreliable results. These usually focus on providing a small pool of operations in heavily constrained search spaces -- usually cell-based neural networks with pre-defined outer-skeletons. In this work, we conducted an empirical analysis of the widely used NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks in terms of their generability and how different operations influence the performance of the generated architectures. We found that only a subset of the operation pool is required to generate architectures close to the upper-bound of the performance range. Also, the performance distribution is negatively skewed, having a higher density of architectures in the upper-bound range. We consistently found convolution layers to have the highest impact on the architecture's performance, and that specific combination of operations favors top-scoring architectures. These findings shed insights on the correct evaluation and comparison of NAS methods using NAS benchmarks, showing that directly searching on NAS-Bench-201, ImageNet16-120 and TransNAS-Bench-101 produces more reliable results than searching only on CIFAR-10. Furthermore, with this work we provide suggestions for future benchmark evaluations and design. The code used to conduct the evaluations is available at https://github.com/VascoLopes/NAS-Benchmark-Evaluation.\",\"PeriodicalId\":13641,\"journal\":{\"name\":\"Inf. Sci.\",\"volume\":\"5 1\",\"pages\":\"119695\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-03-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Inf. Sci.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2303.16938\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Inf. Sci.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2303.16938","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

Neural Architecture Search (NAS)基准测试显著提高了开发和比较NAS方法的能力,同时通过提供关于数千个训练过的神经网络的元信息,大大降低了计算开销。然而,表格基准测试有几个缺点,可能会妨碍公平的比较,并提供不可靠的结果。这些通常侧重于在严重受限的搜索空间中提供一小部分操作——通常是具有预定义外部骨架的基于细胞的神经网络。在这项工作中,我们对广泛使用的NAS-Bench-101、NAS-Bench-201和TransNAS-Bench-101基准进行了实证分析,分析了它们的可通用性,以及不同的操作如何影响所生成架构的性能。我们发现,只需要操作池的一个子集就可以生成接近性能范围上限的体系结构。此外,性能分布呈负偏态,在上限范围内具有更高的架构密度。我们一直发现卷积层对体系结构的性能有最大的影响,并且特定的操作组合有利于得分最高的体系结构。这些发现揭示了使用NAS基准正确评估和比较NAS方法的见解,表明直接在NAS- bench -201、ImageNet16-120和TransNAS-Bench-101上搜索比仅在CIFAR-10上搜索产生更可靠的结果。此外,通过这项工作,我们为未来的基准评估和设计提供了建议。用于进行评估的代码可在https://github.com/VascoLopes/NAS-Benchmark-Evaluation上获得。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Are Neural Architecture Search Benchmarks Well Designed? A Deeper Look Into Operation Importance
Neural Architecture Search (NAS) benchmarks significantly improved the capability of developing and comparing NAS methods while at the same time drastically reduced the computational overhead by providing meta-information about thousands of trained neural networks. However, tabular benchmarks have several drawbacks that can hinder fair comparisons and provide unreliable results. These usually focus on providing a small pool of operations in heavily constrained search spaces -- usually cell-based neural networks with pre-defined outer-skeletons. In this work, we conducted an empirical analysis of the widely used NAS-Bench-101, NAS-Bench-201 and TransNAS-Bench-101 benchmarks in terms of their generability and how different operations influence the performance of the generated architectures. We found that only a subset of the operation pool is required to generate architectures close to the upper-bound of the performance range. Also, the performance distribution is negatively skewed, having a higher density of architectures in the upper-bound range. We consistently found convolution layers to have the highest impact on the architecture's performance, and that specific combination of operations favors top-scoring architectures. These findings shed insights on the correct evaluation and comparison of NAS methods using NAS benchmarks, showing that directly searching on NAS-Bench-201, ImageNet16-120 and TransNAS-Bench-101 produces more reliable results than searching only on CIFAR-10. Furthermore, with this work we provide suggestions for future benchmark evaluations and design. The code used to conduct the evaluations is available at https://github.com/VascoLopes/NAS-Benchmark-Evaluation.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信