ConSense: Continually Sensing Human Activity with WiFi via Growing and Picking

2025-02-18

AI Summary

To address catastrophic forgetting in WiFi-based human activity recognition (HAR) under class-incremental learning—where novel activities are dynamically introduced, edge devices suffer from severe resource constraints, and historical samples cannot be stored—this paper proposes a sample-free class-incremental learning framework. Methodologically, we design a lightweight Transformer-based architecture: parameter-efficient growth is achieved via dynamic expansion of multi-head self-attention layers; selective retraining, guided by neuron stability, and incremental knowledge consolidation jointly preserve prior task performance without retaining old samples. Extensive experiments on three public WiFi HAR benchmarks demonstrate that our method reduces model parameters by 37% and improves average accuracy by 5.2% over state-of-the-art incremental learning approaches, significantly enhancing feasibility and practicality for edge deployment.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Video Understanding & Activity Analysis
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

WiFi-based human activity recognition (HAR) holds significant application potential across various fields. To handle dynamic environments where new activities are continuously introduced, WiFi-based HAR systems must adapt by learning new concepts without forgetting previously learned ones. Furthermore, retaining knowledge from old activities by storing historical exemplar is impractical for WiFi-based HAR due to privacy concerns and limited storage capacity of edge devices. In this work, we propose ConSense, a lightweight and fast-adapted exemplar-free class incremental learning framework for WiFi-based HAR. The framework leverages the transformer architecture and involves dynamic model expansion and selective retraining to preserve previously learned knowledge while integrating new information. Specifically, during incremental sessions, small-scale trainable parameters that are trained specifically on the data of each task are added in the multi-head self-attention layer. In addition, a selective retraining strategy that dynamically adjusts the weights in multilayer perceptron based on the performance stability of neurons across tasks is used. Rather than training the entire model, the proposed strategies of dynamic model expansion and selective retraining reduce the overall computational load while balancing stability on previous tasks and plasticity on new tasks. Evaluation results on three public WiFi datasets demonstrate that ConSense not only outperforms several competitive approaches but also requires fewer parameters, highlighting its practical utility in class-incremental scenarios for HAR.

Problem

• Adapt WiFi-based HAR to new activities
• Retain knowledge without storing historical data
• Reduce computational load with selective retraining

Innovation

• Transformer architecture integration
• Dynamic model expansion strategy
• Selective retraining approach