nanoML for Human Activity Recognition

2025-02-13

AI Summary

To address the energy-efficiency bottleneck of human activity recognition (HAR) on wearable and edge devices, this work proposes a novel nanoML paradigm based on differentiable weightless neural networks (DWNs). It is the first to apply DWNs to HAR—eliminating floating-point operations and parameter storage—and implements ultra-low-power inference on FPGA hardware. Evaluated on standard HAR benchmarks, the approach achieves 96.34%–96.67% accuracy, with per-sample energy consumption of only 56–104 nJ and inference latency as low as 5 ns. Compared to state-of-the-art deep models, it reduces energy consumption by 926,000× and memory footprint by 260×. This work establishes a new technical frontier for nanojoule- and nanosecond-scale HAR, delivering a scalable, ultra-lightweight paradigm for real-time intelligent sensing under extreme resource constraints.

Technology Category

• Machine Learning: Hardware-aware ML
• Humans and AI: Human-in-the-loop Machine Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Human Activity Recognition (HAR) is critical for applications in healthcare, fitness, and IoT, but deploying accurate models on resource-constrained devices remains challenging due to high energy and memory demands. This paper demonstrates the application of Differentiable Weightless Neural Networks (DWNs) to HAR, achieving competitive accuracies of 96.34% and 96.67% while consuming only 56nJ and 104nJ per sample, with an inference time of just 5ns per sample. The DWNs were implemented and evaluated on an FPGA, showcasing their practical feasibility for energy-efficient hardware deployment. DWNs achieve up to 926,000x energy savings and 260x memory reduction compared to state-of-the-art deep learning methods. These results position DWNs as a nano-machine learning nanoML model for HAR, setting a new benchmark in energy efficiency and compactness for edge and wearable devices, paving the way for ultra-efficient edge AI.

Problem

• Energy-efficient HAR on devices
• DWNs reduce memory, energy use
• FPGA implementation for edge AI

Innovation

• DWNs for energy-efficient HAR
• FPGA implementation of DWNs
• Significant energy and memory savings