通过五阶段优化实现AI能效94.6%下降，兼顾性能与环保。
Tu(r)ning AI Green: Exploring Energy Efficiency Cascading with Orthogonal Optimizations
- 按数据、模型、训练、系统、推理五阶段协同优化能效
- 组合优化可降低94.6%能耗，保留95.95%原始F1得分
- 适合关注绿色AI与可持续部署的研究者与工程师
AI的指数级增长加剧了计算需求与能源挑战。尽管从业者采用多种优化手段（本文称其为“调节旋钮”），但这些措施通常作为事后补救，孤立地、被动地应用，缺乏对它们在能效上组合效应的理解。本文强调将能效作为核心设计目标，贯穿计算密集型流程。我们证明，在五个AI流程阶段（数据、模型、训练、系统、推理）中进行策略性选择，可产生级联式效率提升。实验验证表明，正交组合可使能耗最多降低94.6%，同时保持非优化流程95.95%的原始F1分数。该系统化方法为可持续AI提供了可操作的框架，平衡效率、性能与环境责任。
原文摘要 · Abstract (English)
AI's exponential growth intensifies computational demands and energy challenges. While practitioners employ various optimization techniques, that we refer as "knobs" in this paper, to tune model efficiency, these are typically afterthoughts and reactive ad-hoc changes applied in isolation without understanding their combinatorial effects on energy efficiency. This paper emphasizes on treating energy efficiency as the first-class citizen and as a fundamental design consideration for a compute-intensive pipeline. We show that strategic selection across five AI pipeline phases (data, model, training, system, inference) creates cascading efficiency. Experimental validation shows orthogonal combinations reduce energy consumption by up to $94.6$% while preserving $95.95$% of the original F1 score of non-optimized pipelines. This curated approach provides actionable frameworks for informed sustainable AI that balance efficiency, performance, and environmental responsibility.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。