• Breadcrumb Home News Energy-aware Machine Learning Energy-aware machine learning New program aims to strike balance between energy usage and performance Traditional machine learning models are designed with a single focus: maximizing performance. • While this approach has driven breakthroughs in language models, image recognition, and other domains, it has overlooked a critical factor â electricity consumption. • Our Mapping Machine Learning to Physics (ML2P) programis designed to transform how artificial intelligence (AI) systems balance performance with energy use. • ML2P addresses this gap by mapping machine learning model performance to physical electric characteristics, with a focus on measuring energy use in joules. • By embedding energy-awareness into the design of AI systems, ML2P aims to create models that can achieve the right balance between accuracy and power consumption. • â In an era where AI is increasingly deployed in power-constrained environments, such as at the tactical edge, energy efficiency is no longer optional,â said Bernard McShea, founding program manager for ML2P.
Article Summaries:
- DARPA has launched the Mapping Machine Learning to Physics (ML2P) program to integrate energy consumption into AI design. The initiative seeks to balance model accuracy with power use by mapping performance to physical electric characteristics measured in joules. Experts from electrical engineering, mathematics, logic, and machine learning will develop training functions that optimize this trade‑off, targeting applications in power‑constrained environments such as tactical edge systems. Beyond defense, ML2P aims to improve the energy efficiency of large language models, generative AI, and classification tasks on existing hardware and inform future AI‑optimized hardware design. Researchers can learn more or submit proposals via SAM.gov.
Sources: