EdgeSync: Accelerating Edge-Model Updates for Data Drift through Adaptive Continuous Learning

TL;DR

EdgeSync is a novel framework that accelerates and improves the relevance of model updates for edge devices in real-time video analytics, effectively handling data drift with reduced delay and increased accuracy.

Contribution

It introduces an adaptive sample filtering and dynamic training management approach to enhance the efficiency and relevance of edge-model updates under data drift conditions.

Findings

Improves accuracy by approximately 3.4% over existing methods.

Reduces update delays through efficient sample filtering.

Enhances model relevance with dynamic training timing.

Abstract

Real-time video analytics systems typically deploy lightweight models on edge devices to reduce latency. However, the distribution of data features may change over time due to various factors such as changing lighting and weather conditions, leading to decreased model accuracy. Recent frameworks try to address this issue by leveraging remote servers to continuously train and adapt lightweight edge models using more complex models in the cloud. Despite these advancements, existing methods face two key challenges: first, the retraining process is compute-intensive, causing significant delays in model updates; second, the new model may not align well with the evolving data distribution of the current video stream. To address these challenges, we introduce EdgeSync, an efficient edge-model updating approach that enhances sample filtering by incorporating timeliness and inference results, thus ensuring training samples are more relevant to the current video content while reducing update delays. Additionally, EdgeSync features a dynamic training management module that optimizes the timing and sequencing of model updates to improve their timeliness. Evaluations on diverse and complex real-world datasets demonstrate that EdgeSync improves accuracy by approximately 3.4% compared to existing methods and by about 10% compared to traditional approaches.