New method optimizes DNN inference for energy efficiency on mobile devices

Researchers have developed a new method for optimizing energy efficiency in deep neural network (DNN) inference on mobile devices. The approach jointly optimizes both computing and memory frequencies, along with communication resources, to minimize energy consumption while adhering to deadline constraints. For local inference, a near-optimal closed-form solution was derived using convex optimization, achieving performance within 2.5% of the optimal solution and reducing energy consumption by up to 10.4% compared to existing methods. AI

IMPACT This research could lead to more energy-efficient AI applications on mobile devices, extending battery life and enabling more complex on-device processing.

RANK_REASON The cluster contains an academic paper detailing a new method for optimizing DNN inference. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• deep neural network
• dynamic voltage and frequency scaling
• Hugging Face

• paper
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →