Device Learning

Device learning focuses on training and deploying machine learning models directly on resource-constrained devices like smartphones and microcontrollers, aiming to improve efficiency, privacy, and responsiveness. Current research emphasizes developing lightweight algorithms (e.g., sparse backpropagation, binary neural networks) and architectures (e.g., EEGNet, ResNet) optimized for low-power hardware, often incorporating techniques like self-supervised learning, federated learning, and continual learning to address data limitations and dynamic environments. This field is significant for enabling intelligent functionalities in edge devices across diverse applications, from wearable health monitoring to autonomous robots, while minimizing reliance on cloud infrastructure and preserving data privacy.

Papers

August 29, 2023

On-Device Learning with Binary Neural Networks
Lorenzo Vorabbi, Davide Maltoni, Stefano Santi
Deep Learning Model Continual LEArning Binary Neural Network Device Learning Mixed Quantization Binary Network

June 15, 2023

MobileASR: A resource-aware on-device learning framework for user voice personalization applications on mobile phones
Zitha Sasindran, Harsha Yelchuri, Pooja Rao, T. V. Prabhakar
Automatic Speech Recognition Speech Recognition Efficient Training Automatic Speech Recognition Model Training Model Device Training Device Learning Mobile Phone

May 30, 2023

Reduced Precision Floating-Point Optimization for Deep Neural Network On-Device Learning on MicroControllers
Davide Nadalini, Manuele Rusci, Luca Benini, Francesco Conti
Low Precision Device Learning 32 Bit Microcontrollers Device Deep Floating Point

May 4, 2023

CAMEL: Co-Designing AI Models and Embedded DRAMs for Efficient On-Device Learning
Sai Qian Zhang, Thierry Tambe, Nestor Cuevas, Gu-Yeon Wei, David Brooks
Device Training Device Learning Electric SHEEP Human AI Co Creation Random Access Memory Chip Memory 1T1C DRAM Cell

January 17, 2023

Async-HFL: Efficient and Robust Asynchronous Federated Learning in Hierarchical IoT Networks
Xiaofan Yu, Ludmila Cherkasova, Harsh Vardhan, Quanling Zhao, Emily Ekaireb, Xiyuan Zhang, Arya Mazumdar, Tajana Rosing
Internet of Thing Internet of Thing Network Asynchronous Federated Learning Device Learning

December 19, 2022

A Sequential Concept Drift Detection Method for On-Device Learning on Low-End Edge Devices
Takeya Yamada, Hiroki Matsutani
Edge Device Concept Drift Device Learning Concept Drift Detection

October 12, 2022

VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation
Megh Shukla, Roshan Roy, Pankaj Singh, Shuaib Ahmed, Alexandre Alahi
Active Learning Distribution Detection Pose Estimation Human Pose Device Learning Pose Refinement

August 25, 2022

Enabling Weakly-Supervised Temporal Action Localization from On-Device Learning of the Video Stream
Yue Tang, Yawen Wu, Peipei Zhou, Jingtong Hu
Temporal Action Localization Action Detection Device Learning Device Use Case Video Stream Efficient Video

August 1, 2022

Efficient Personalized Learning for Wearable Health Applications using HyperDimensional Computing
Sina Shahhosseini, Yang Ni, Hamidreza Alikhani, Emad Kasaeyan Naeini, Mohsen Imani, Nikil Dutt, Amir M. Rahmani
Wearable Device Hyperdimensional Computing Health Monitoring Personalized Learning Device Learning Wearable Data Collection System

June 22, 2022

Quantization Robust Federated Learning for Efficient Inference on Heterogeneous Devices
Kartik Gupta, Marios Fournarakis, Matthias Reisser, Christos Louizos, Markus Nagel
Efficient Inference Federated Averaging Device Learning Federated Optimization Heterogeneous Device

June 7, 2022

FEL: High Capacity Learning for Recommendation and Ranking via Federated Ensemble Learning
Meisam Hejazinia, Dzmitry Huba, Ilias Leontiadis, Kiwan Maeng, Mani Malek, Luca Melis, Ilya Mironov, Milad Nasr, Kaikai Wang, Carole-Jean Wu
Effective Recommendation Partial Ranking Deep Learning Application Device Training Device Learning Federated Ensemble Knowledge Capacity

May 3, 2022

Efficient Fine-Tuning of BERT Models on the Edge
Danilo Vucetic, Mohammadreza Tayaranian, Maryam Ziaeefard, James J. Clark, Brett H. Meyer, Warren J. Gross
Extreme Edge BERT Based Device Training Pre Trained BERT Device Learning Memory Efficient Training

March 21, 2022

Online Continual Learning for Embedded Devices
Tyler L. Hayes, Christopher Kanan
Continual LEArning Continual Learning Online Continual Learning Embedded System Device Learning Continual Learning Model

March 2, 2022

Addressing Gap between Training Data and Deployed Environment by On-Device Learning
Kazuki Sunaga, Masaaki Kondo, Hiroki Matsutani
Training Data Edge Device Filling Gap TinyML Model Device Learning Sensor Node

February 14, 2022

Federated Contrastive Learning for Dermatological Disease Diagnosis via On-device Learning
Yawen Wu, Dewen Zeng, Zhepeng Wang, Yi Sheng, Lei Yang, Alaina J. James, Yiyu Shi, Jingtong Hu
Contrastive Learning Deep Learning Model Device Learning Dermatological Disease Diagnosis Federated Contrastive Learning Dermatological Disease Datasets

January 24, 2022

On-Device Learning with Cloud-Coordinated Data Augmentation for Extreme Model Personalization in Recommender Systems
Renjie Gu, Chaoyue Niu, Yikai Yan, Fan Wu, Shaojie Tang, Rongfeng Jia, Chengfei Lyu, Guihai Chen
Data Augmentation Recommender System Recommendation Model Device Training Device Learning Model Personalization Cloud Collaborative Learning Collaborative Edge Training

December 16, 2021

DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systems
Martin Rapp, Ramin Khalili, Kilian Pfeiffer, Jörg Henkel
Neural Network Convolutional Layer Adaptive Learning Device Learning Heterogeneous System Dropout Layer

Device Learning

Papers

On-Device Learning with Binary Neural Networks

MobileASR: A resource-aware on-device learning framework for user voice personalization applications on mobile phones

Reduced Precision Floating-Point Optimization for Deep Neural Network On-Device Learning on MicroControllers

CAMEL: Co-Designing AI Models and Embedded DRAMs for Efficient On-Device Learning

Async-HFL: Efficient and Robust Asynchronous Federated Learning in Hierarchical IoT Networks

A Sequential Concept Drift Detection Method for On-Device Learning on Low-End Edge Devices

VL4Pose: Active Learning Through Out-Of-Distribution Detection For Pose Estimation

Enabling Weakly-Supervised Temporal Action Localization from On-Device Learning of the Video Stream

Efficient Personalized Learning for Wearable Health Applications using HyperDimensional Computing

Quantization Robust Federated Learning for Efficient Inference on Heterogeneous Devices

FEL: High Capacity Learning for Recommendation and Ranking via Federated Ensemble Learning

Efficient Fine-Tuning of BERT Models on the Edge

Online Continual Learning for Embedded Devices

Addressing Gap between Training Data and Deployed Environment by On-Device Learning

Federated Contrastive Learning for Dermatological Disease Diagnosis via On-device Learning

On-Device Learning with Cloud-Coordinated Data Augmentation for Extreme Model Personalization in Recommender Systems

DISTREAL: Distributed Resource-Aware Learning in Heterogeneous Systems