Prism: Mining Task-aware Domains in Non-i.i.d. IMU Data for Flexible User Perception

2025-01-03

AI Summary

Flexible user perception (FUP) suffers from poor generalization on mobile devices due to non-i.i.d. IMU data across users and postures. Method: We propose a task-driven implicit domain discovery framework—the first to formulate FUP as a joint domain identification and adaptive model selection problem. Leveraging an EM algorithm tailored to the perception task, our approach automatically discovers semantically coherent implicit domains and assigns lightweight, domain-specific models. It enables dynamic sample-to-domain matching and low-latency inference, augmented by a feature-space domain similarity metric for unconstrained online prediction. Results: Evaluated on multiple smartphones, our method achieves inference latency <20 ms and improves accuracy by 12.7–28.3% in cross-user and cross-posture scenarios, significantly outperforming state-of-the-art approaches.

Technology Category

• Application Domains: Mobility, Driving & Flight
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

A wide range of user perception applications leverage inertial measurement unit (IMU) data for online prediction. However, restricted by the non-i.i.d. nature of IMU data collected from mobile devices, most systems work well only in a controlled setting (e.g., for a specific user in particular postures), limiting application scenarios. To achieve uncontrolled online prediction on mobile devices, referred to as the flexible user perception (FUP) problem, is attractive but hard. In this paper, we propose a novel scheme, called Prism, which can obtain high FUP accuracy on mobile devices. The core of Prism is to discover task-aware domains embedded in IMU dataset, and to train a domain-aware model on each identified domain. To this end, we design an expectation-maximization (EM) algorithm to estimate latent domains with respect to the specific downstream perception task. Finally, the best-fit model can be automatically selected for use by comparing the test sample and all identified domains in the feature space. We implement Prism on various mobile devices and conduct extensive experiments. Results demonstrate that Prism can achieve the best FUP performance with a low latency.

Problem

• Inertial Measurement Unit (IMU)
• Flexible User Perception (FUP)
• Unrestricted Online Prediction

Innovation

• Prism
• Flexible User Perception
• Expectation Maximization Algorithm