Article

Efficient and Secure<i>μ</i>-Training and<i>μ</i>-Fine-Tuning for Edge-Based TinyML with Future-Guided Self-Distillation

2025-02-04

Abstract excerpt

This study presents a novel, computationally efficient training framework demonstrated through bio-signal processing on edge medical devices. The approach integrates conventional full training with an innovative µ -Training technique, wherein the encoder and decoder of a compact model remain frozen while only the middle layer is updated. This design is further enhanced by a novel Future-Guided Self-Distillation me...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 96944a21-310d-53d7-af42-5c508562a4da
DOI: 10.1101/2025.01.30.25321374

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Efficient and Secure<i>μ</i>-Training and<i>μ</i>-Fine-Tuning for Edge-Based TinyML with Future-Guided Self-DistillationDOI 10.1101/2025.01.30.25321374Read

Lightweight Transformer Models for Biomedical Signal Processing: Trends, Challenges, and Future DirectionsSemantic proximity Read

CardioFM: A Multimodal Foundation Model for Joint ECG and PPG Representation LearningSemantic proximity Read

ECG-SSL: A Systematic Benchmark for Self-SupervisedLearning Strategies on Multi-Center Electrocardiogram DataSemantic proximity Read

Lightweight Deep Learning for Atrial Fibrillation Detection from Single-Lead Wearable ECG: A Systematic Benchmark of Convolutional, Temporal, and State-Space ArchitecturesSemantic proximity Read

Cardiac abnormality detection with a tiny diagonal state space model based on sequential liquid neural processing unitsSemantic proximity Read

Timesynth: A Temporal Fidelity Framework for Health Signal Digital TwinsSemantic proximity Read

Human-Centered Learning Mechanics for Reliable Smart Textiles: An Explainable and Frugal AI Framework for Separating Physiological Anomalies from Sensor Drift and DegradationSemantic proximity Read

Towards Personalized Edge-AI for Medicine: Efficient Neuromorphic Frameworks for Seizure Detection and PredictionSemantic proximity Read

Smartwatch Photoplethysmography-Derived Heart Age via ECG-Guided Cross-Modal Pretraining as a Digital Biomarker of Vascular AgingSemantic proximity Read

NDPAI: Neuroscience-Derived Predictive Active Inference for Calibrated Cross-Subject Wearable Stress DetectionSemantic proximity Read

HybridNet-XR: Efficient Teacher-Free Self-Supervised Learning for Autonomous Medical Diagnostic Systems in Resource-Constrained EnvironmentsSemantic proximity Read

PIPxResNet: Penalty Induced Prototype-Based eXplainable Residual Neural Network for Heartbeat ClassificationSemantic proximity Read