Enhancing User Experience in On-Device Machine Learning with Gated Compression Layers

arXiv:2405.01739v1 Announce Type: new
Abstract: On-device machine learning (ODML) enables powerful edge applications, but power consumption remains a key challenge for resource-constrained devices. To address this, developers often face a trade-off between model accuracy and power consumption, employing either computationally intensive models on high-power cores or pared-down models on low-power cores. Both approaches typically lead to a compromise in user experience (UX). This work focuses on the use of Gated Compression (GC) layer to enhance ODML model performance while conserving …

abstract accuracy applications arxiv challenge compression consumption cs.lg developers devices edge experience face key machine machine learning model accuracy power power consumption trade trade-off type