Characterizing the Energy Requirement of Computer Vision

Abstract

The energy requirements of neural network learning are growing at a rapid rate. Increased energy demands have caused a global need to seek ways to improve energy efficiency of neural network learning. This thesis aims to establish a baseline on how adjusting basic parameters can affect energy consumption in neural network learning on Computer Vision tasks. I catalogued the effects of various adjust adjustment from simple batch size adjustment to more complicated hardware configuration (such as power capping). Findings include that adjusting from single precision model to a mixed precision model can result in energy reductions of nearly 40%. Additionally power capping the GPU can reduce energy cost by an additional 10%.