Tu(r)ning AI Green: Exploring Energy Efficiency Cascading with Orthogonal Optimizations
This addresses the environmental impact of AI for practitioners by offering a framework to reduce energy usage without significant performance loss, though it is incremental in combining existing optimizations.
The paper tackles the problem of AI's high energy consumption by proposing a strategic, holistic approach to energy efficiency across five pipeline phases, achieving up to 94.6% reduction in energy use while maintaining 95.95% of the original F1 score.
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.