Addressing Data Distribution Shifts in Online Machine Learning Powered Smart City Applications Using Augmented Test-Time Adaptation

TL;DR

This paper introduces a systematic active fine-tuning layer for online machine learning in smart city applications, effectively addressing data distribution shifts at test time with improved performance and cost-efficiency.

Contribution

It proposes a novel SAF layer that enables continuous, intelligent, and cost-effective test-time adaptation in smart city machine learning models.

Findings

Outperforms traditional test-time adaptation by a factor of two

Enhances robustness to data distribution shifts in smart city data

Reduces relabeling costs through budgeted human-machine collaboration

Abstract

Data distribution shift is a common problem in machine learning-powered smart city applications where the test data differs from the training data. Augmenting smart city applications with online machine learning models can handle this issue at test time, albeit with high cost and unreliable performance. To overcome this limitation, we propose to endow test-time adaptation with a systematic active fine-tuning (SAF) layer that is characterized by three key aspects: a continuity aspect that adapts to ever-present data distribution shifts; intelligence aspect that recognizes the importance of fine-tuning as a distribution-shift-aware process that occurs at the appropriate time to address the recently detected data distribution shifts; and cost-effectiveness aspect that involves budgeted human-machine collaboration to make relabeling cost-effective and practical for diverse smart city applications. Our empirical results show that our proposed approach outperforms the traditional test-time adaptation by a factor of two.