FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors

TL;DR

Flame is a self-adaptive auto-labeling system designed for heterogeneous mobile processors, enabling accurate and efficient labeling of non-stationary data with unknown labels on mobile devices.

Contribution

This paper introduces Flame, a novel runtime system that efficiently schedules and executes auto-labeling workloads on diverse mobile hardware, handling non-stationary data and unknown labels.

Findings

Achieves high labeling accuracy on mobile devices.

Demonstrates high performance across eight datasets.

Effectively manages hardware heterogeneity for auto-labeling.

Abstract

How to accurately and efficiently label data on a mobile device is critical for the success of training machine learning models on mobile devices. Auto-labeling data on mobile devices is challenging, because data is usually incrementally generated and there is possibility of having unknown labels. Furthermore, the rich hardware heterogeneity on mobile devices creates challenges on efficiently executing auto-labeling workloads. In this paper, we introduce Flame, an auto-labeling system that can label non-stationary data with unknown labels. Flame includes a runtime system that efficiently schedules and executes auto-labeling workloads on heterogeneous mobile processors. Evaluating Flame with eight datasets on a smartphone, we demonstrate that Flame enables auto-labeling with high labeling accuracy and high performance.