Constrained Device

Constrained devices, characterized by limited computational resources and memory, are the focus of intense research aimed at enabling powerful machine learning applications in resource-scarce environments. Current efforts concentrate on adapting existing deep learning models, such as CNNs and specialized architectures like EfficientNet, through techniques like pruning, quantization, and layer-wise training in federated learning settings to improve efficiency and accuracy on these devices. This research is crucial for expanding the reach of AI to diverse applications, including healthcare diagnostics in remote areas, predictive maintenance of infrastructure like solar farms, and enabling intelligent functionalities in resource-constrained IoT devices.

Papers

August 15, 2024

Moving Healthcare AI-Support Systems for Visually Detectable Diseases onto Constrained Devices
Tess Watt, Christos Chrysoulas, Peter J Barclay
Image Classification Disease Detection TinyML Model Medical AI AI Assistant Communication Limited Environment Constrained Device

March 8, 2024

Micro-Fracture Detection in Photovoltaic Cells with Hardware-Constrained Devices and Computer Vision
Booy Vitas Faassen, Jorge Serrano, Paul D. Rosero-Montalvo
Computer Vision Fracture Detection Solar Cell Renewable Energy Source Silicon Atom Constrained Device

May 26, 2023

Aggregating Capacity in FL through Successive Layer Training for Computationally-Constrained Devices
Kilian Pfeiffer, Ramin Khalili, Jörg Henkel
Federated Learning Edge Device Capacity Loss Constrained Device Co Adaptation Successive Affine Learning

March 18, 2023

Weight-sharing Supernet for Searching Specialized Acoustic Event Classification Networks Across Device Constraints
Guan-Ting Lin, Qingming Tang, Chieh-Chi Kao, Viktor Rozgic, Chao Wang
Neural Architecture Search Event Classification Supernet Training Smart Speaker Weight Sharing Supernet Constrained Device

November 8, 2022

Enhancing Efficiency in Multidevice Federated Learning through Data Selection
Fan Mo, Mohammad Malekzadeh, Soumyajit Chatterjee, Fahim Kawsar, Akhil Mathur
Data Selection Resource Constrained Device Enhancing Efficiency Cross Device Federated Learning Benchmarking Deep Learning Constrained Device

May 26, 2022

FCN-Pose: A Pruned and Quantized CNN for Robot Pose Estimation for Constrained Devices
Marrone Silvério Melo Dantas, Iago Richard Rodrigues, Assis Tiago Oliveira Filho, Gibson Barbosa, Daniel Bezerra, Djamel F. H. Sadok, Judith Kelner, Maria Marquezini, Ricardo Silva
Deep Learning Pose Estimation Edge Pruning Deep Learning Application Stable Pose Robot Pose Constrained Device

March 10, 2022

CoCoFL: Communication- and Computation-Aware Federated Learning via Partial NN Freezing and Quantization
Kilian Pfeiffer, Martin Rapp, Ramin Khalili, Jörg Henkel
Federated Learning Quantization Operator Timely Communication Heterogeneous Source Frozen Pre Trained Accuracy Disparity Constrained Device

Constrained Device

Papers

Moving Healthcare AI-Support Systems for Visually Detectable Diseases onto Constrained Devices

Micro-Fracture Detection in Photovoltaic Cells with Hardware-Constrained Devices and Computer Vision

Aggregating Capacity in FL through Successive Layer Training for Computationally-Constrained Devices

Weight-sharing Supernet for Searching Specialized Acoustic Event Classification Networks Across Device Constraints

Enhancing Efficiency in Multidevice Federated Learning through Data Selection

FCN-Pose: A Pruned and Quantized CNN for Robot Pose Estimation for Constrained Devices

CoCoFL: Communication- and Computation-Aware Federated Learning via Partial NN Freezing and Quantization