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 2402.12010cs.LGcs.AIcs.NE
收藏

使用精英样本训练绿色AI模型

Training Green AI Models Using Elite Samples

University of Luxembourg(卢森堡大学);University of Florence(佛罗伦萨大学)

Mohammed Alswaitti, Roberto Verdecchia, Grégoire Danoy, Pascal Bouvry, Johnatan Pecero

AI总结:

本文提出一种基于进化的采样框架,通过识别精英训练样本来减少训练集规模,实验证明仅用10%样本即可提升50%性能并节省98%能源,促进可持续AI训练。

摘要:

AI模型训练的大幅增加带来了相当大的环境影响,要求采用更节能和可持续的AI实践。一方面,以数据为中心的方法在训练节能AI模型方面展现出巨大潜力。另一方面,实例选择方法证明了在最小化训练集且性能几乎不下降的情况下训练AI模型的能力。尽管这两个主题日益受到关注,但以数据为中心的训练集选择对能效的影响至今仍未被探索。本文提出了一个基于进化的采样框架,旨在(i)识别针对数据集和模型对量身定制的精英训练样本,(ii)将模型性能和能效提升与典型模型训练实践进行比较,以及(iii)研究该框架在促进可持续模型训练实践方面的可行性。为了评估所提出的框架,我们进行了一项实证实验,包括8个常用的AI分类模型和25个公开可用的数据集。结果表明,通过考虑10%的精英训练样本,与常规训练实践相比,模型性能可提升50%,并显著节省98%的能源。

英文摘要:

The substantial increase in AI model training has considerable environmental implications, mandating more energy-efficient and sustainable AI practices. On the one hand, data-centric approaches show great potential towards training energy-efficient AI models. On the other hand, instance selection methods demonstrate the capability of training AI models with minimised training sets and negligible performance degradation. Despite the growing interest in both topics, the impact of data-centric training set selection on energy efficiency remains to date unexplored. This paper presents an evolutionary-based sampling framework aimed at (i) identifying elite training samples tailored for datasets and model pairs, (ii) comparing model performance and energy efficiency gains against typical model training practice, and (iii) investigating the feasibility of this framework for fostering sustainable model training practices. To evaluate the proposed framework, we conducted an empirical experiment including 8 commonly used AI classification models and 25 publicly available datasets. The results showcase that by considering 10% elite training samples, the models' performance can show a 50% improvement and remarkable energy savings of 98% compared to the common training practice.

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。

↑