SemiPFL: Personalized Semi-Supervised Federated Learning Framework for Edge Intelligence

Published in

IEEE Internet of Things Journal(2022)

External Links:

Generate Graph DownloadPDF

TL;DR

It is demonstrated that SemiPFL outperforms state-of-the-art federated learning frameworks under the same assumptions regarding user performance, network footprint, and computational consumption, and the solution performs well for users without label or having limited labeled data sets and increasing performance for increased labeled data and number of users.

Abstract

Recent advances in wearable devices and Internet of Things (IoT) have led to massive growth in sensor data generated in edge devices. Labeling such massive data for classification tasks has proven to be challenging. In addition, data generated by different users bear various personal attributes and edge heterogeneity, rendering it impractical to develop a global model that adapts well to all users. Concerns over data privacy and communication costs also prohibit centralized data accumulation and training. We propose SemiPFL that supports edge users having no label or limited labeled data sets and a sizable amount of unlabeled data that is insufficient to train a well-performing model. In this work, edge users collaborate to train a Hyper-network in the server, generating personalized autoencoders for each user. After receiving updates from edge users, the server produces a set of base models for each user, which the users locally aggregate them using their own labeled data set. We comprehensively evaluate our proposed framework on various public data sets from a wide range of application scenarios, from wearable health to IoT, and demonstrate that SemiPFL outperforms state-of-the-art federated learning frameworks under the same assumptions regarding user performance, network footprint, and computational consumption. We also show that the solution performs well for users without label or having limited labeled data sets and increasing performance for increased labeled data and number of users, signifying the effectiveness of SemiPFL for handling data heterogeneity and limited annotation. We also demonstrate the stability of SemiPFL for handling user hardware resource heterogeneity in three real-time scenarios.

SemiPFL: Personalized Semi-Supervised Federated Learning Framework for Edge Intelligence

Published in

IEEE Internet of Things Journal(2022)

External Links:

Generate Graph DownloadPDF

TL;DR

Abstract

References83 items

Retraction notice to 'Personalized Federated Learning Framework for Network Traffic Anomaly Detection'

FedHAR: Semi-Supervised Online Learning for Personalized Federated Human Activity Recognition

Federated Learning with Label Distribution Skew via Logits Calibration

Learn from Others and Be Yourself in Heterogeneous Federated Learning

Deep CNN-LSTM With Self-Attention Model for Human Activity Recognition Using Wearable Sensor

FLAME

Data Heterogeneity-Robust Federated Learning via Group Client Selection in Industrial IoT

Federated Learning Challenges and Opportunities: An Outlook

Dynamic Contract Design for Federated Learning in Smart Healthcare Applications

Local Learning Matters: Rethinking Data Heterogeneity in Federated Learning

Federated learning for predicting clinical outcomes in patients with COVID-19

FedRS: Federated Learning with Restricted Softmax for Label Distribution Non-IID Data

New Metrics to Evaluate the Performance and Fairness of Personalized Federated Learning

A Survey on Federated Learning for Resource-Constrained IoT Devices

Personalized Federated Learning with Gaussian Processes

FedHealth 2: Weighted Federated Transfer Learning via Batch Normalization for Personalized Healthcare

FogFL: Fog-Assisted Federated Learning for Resource-Constrained IoT Devices

Federated Learning for Internet of Things: A Comprehensive Survey

Innovative chest X-ray image recognition technique and its economic value

Federated Learning Meets Human Emotions: A Decentralized Framework for Human–Computer Interaction for IoT Applications

Blockchain-Based Federated Learning for Device Failure Detection in Industrial IoT

Personalized Federated Learning using Hypernetworks

Towards Personalized Federated Learning

Efficient Federated Learning Algorithm for Resource Allocation in Wireless IoT Networks

FedBN: Federated Learning on Non-IID Features via Local Batch Normalization

SelfHAR

A Resource-Constrained and Privacy-Preserving Edge-Computing-Enabled Clinical Decision System: A Federated Reinforcement Learning Approach

Ditto: Fair and Robust Federated Learning Through Personalization

Semi-supervised Federated Learning for Activity Recognition

When Deep Reinforcement Learning Meets Federated Learning: Intelligent Multitimescale Resource Management for Multiaccess Edge Computing in 5G Ultradense Network

Dynamic-Fusion-Based Federated Learning for COVID-19 Detection

Federated Learning: A Survey on Enabling Technologies, Protocols, and Applications

Federated Learning in the Sky: Aerial-Ground Air Quality Sensing Framework With UAV Swarms

Deep Anomaly Detection for Time-Series Data in Industrial IoT: A Communication-Efficient On-Device Federated Learning Approach

Federated Self-Supervised Learning of Multisensor Representations for Embedded Intelligence

Secure, privacy-preserving and federated machine learning in medical imaging

Personalized Federated Learning with Moreau Envelopes

Personalized Federated Learning With Differential Privacy

Multiagent DDPG-Based Deep Learning for Smart Ocean Federated Learning IoT Networks

Privacy-Preserving Traffic Flow Prediction: A Federated Learning Approach

Personalized Federated Learning for Intelligent IoT Applications: A Cloud-Edge Based Framework

Personalized Federated Learning: A Meta-Learning Approach

Advances and Open Problems in Federated Learning

Federated Learning with Personalization Layers

A survey on semi-supervised learning

Federated Learning for Healthcare Informatics

Federated Learning in Mobile Edge Networks: A Comprehensive Survey

Smartphone and Smartwatch-Based Biometrics Using Activities of Daily Living

FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare

CMFL: Mitigating Communication Overhead for Federated Learning

Towards Federated Learning at Scale: System Design

Federated Optimization in Heterogeneous Networks

Introducing WESAD, a Multimodal Dataset for Wearable Stress and Affect Detection

A Survey on Behavior Recognition Using WiFi Channel State Information

Deep Learning for Sensor-based Activity Recognition: A Survey

DeepSleepNet: A Model for Automatic Sleep Stage Scoring Based on Raw Single-Channel EEG

HyperNetworks

Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables

Human Daily Activity and Fall Recognition Using a Smartphone's Acceleration Sensor

Communication-Efficient Learning of Deep Networks from Decentralized Data

Smart Devices are Different: Assessing and MitigatingMobile Sensing Heterogeneities for Activity Recognition

Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition

A Survey on Human Activity Recognition using Wearable Sensors

Sensor-Based Activity Recognition

Introducing a New Benchmarked Dataset for Activity Monitoring

Creating and benchmarking a new dataset for physical activity monitoring

Activity Recognition from Accelerometer Data

Semisupervised and personalized federated activity recognition based on active learning and label propagation

Federated Deep Learning for Cyber Security in the Internet of Things: Concepts, Applications, and Experimental Analysis

Personalized Federated Learning with Theoretical Guarantees: A Model-Agnostic Meta-Learning Approach

Federated Learning for UAVs-Enabled Wireless Networks: Use Cases, Challenges, and Open Problems

Deep Learning

A Public Domain Dataset for Human Activity Recognition using Smartphones

A survey on fall detection: Principles and approaches

Human Activity Recognition and Pattern Discovery

Learning Multiple Layers of Features from Tiny Images

CIFAR-10 (Canadian institute for advanced research)

Her research interests include statistical signal processing and machine learning, with applications in digital media and biomedical data analytics

Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol

Expert Systems With Applications

The proposed method outperforms recognition accuracy compared to recent publications using fully connected neural network architectures

To our knowledge, this is the ﬁrst work that studies the effect of hardware resource heterogeneity among users

To our knowledge, this is the ﬁrst work developing personalized models using a Hyper-network for multi-sensory classiﬁcation