Re-thinking Efficiency in Deep Learning under Accelerators and commodity and processors

Informazioni relative al progetto

REDIAL

ID dell’accordo di sovvenzione: 805194

DOI

10.3030/805194

Progetto chiuso

Data della firma CE 10 Ottobre 2019

Data di avvio 1 Settembre 2020

Data di completamento 31 Agosto 2025

Finanziato da

• EXCELLENT SCIENCE - European Research Council (ERC)

Costo totale

€ 1 495 036,00

Contributo UE

€ 1 495 036,00

Coordinato da

THE CHANCELLOR MASTERS AND SCHOLARS OF THE UNIVERSITY OF CAMBRIDGE
United Kingdom

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Pubblicazioni

Conference proceedings (15)

Secure aggregation for federated learning in flower

Autori: Kwing Hei Li; Pedro Porto Buarque de Gusmão; Daniel J. Beutel; Nicholas D. Lane
Pubblicato in: Proceedings of the 2nd ACM International Workshop on Distributed Machine Learning, Numero 1, 2021, Pagina/e 8-14
Editore: ACM
DOI: 10.48550/arxiv.2205.06117

PROSPECT PRUNING: FINDING TRAINABLE WEIGHTSAT INITIALIZATION USING META-GRADIENTS

Autori: Milad Alizadeh, Shyam A. Tailor, Luisa Zintgraf 3 Joost van Amersfoort 1 Sebastian Farquhar 1 Nicholas Donald Lane 2,4 Yarin Gal 1
Pubblicato in: International Conference on Learning Representations, Numero 1, 2022
Editore: ICLR

Differentiable Neural Network Pruning to Enable Smart Applications on Microcontrollers

Autori: Edgar Liberis, Nicholas D Lane
Pubblicato in: 2023, ISSN 2474-9567
Editore: ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies IMWUT
DOI: 10.48550/arxiv.2110.08350

Federated Self-supervised Speech Representations: Are We There Yet?

Autori: Yan Gao, Javier Fernandez-Marques, Titouan Parcollet, Abhinav Mehrotra, Nicholas D. Lane
Pubblicato in: 2022
Editore: Interspeech 2022

Di!erentiable Neural Network Pruning to Enable Smart Applicationson Microcontrollers

Autori: EDGAR LIBERIS, NICHOLAS D. LANE
Pubblicato in: ACM IMWUT Journal, Numero 1, 2022
Editore: ACM

Deep learning on microcontrollers

Autori: Svoboda, Filip; Fernandez-Marques, Javier; Liberis, Edgar; Lane, Nicholas D.
Pubblicato in: isbn: 9781450392549, Numero 1, 2022
Editore: 2nd Workshop on Machine Learning and Systems, EuroMLSys

Gradient-less Federated Gradient Boosting Tree with Learnable Learning Rates

Autori: Chenyang Ma, Xinchi Qiu, Daniel Beutel, Nicholas Lane
Pubblicato in: 2023
Editore: 3rd Workshop on Machine Learning and Systems

Federated Self-supervised Learning for Video Understanding

Autori: Yasar Abbas Ur Rehman, Yan Gao, Jiajun Shen, Pedro Porto Buarque de Gusmão, Nicholas Lane
Pubblicato in: 2022
Editore: 17th European Conference on Computer Vision

ZeroFL: Efficient On-Device Training for Federated Learning with Local Sparsity

Autori: Qiu, Xinchi; Fernandez-Marques, Javier; Gusmao, Pedro PB; Gao, Yan; Parcollet, Titouan; Lane, Nicholas Donald
Pubblicato in: 2022
Editore: International Conference on Learning Representations
DOI: 10.48550/arxiv.2208.02507

End-to-End Speech Recognition from Federated Acoustic Models

Autori: Yan Gao, Titouan Parcollet, Salah Zaiem, Javier Fernandez-Marques, Pedro P. B. de Gusmao, Daniel J. Beutel, Nicholas Lane
Pubblicato in: 2022
Editore: IEEE International Conference on Acoustics, Speech and Signal Processing

μNAS: Constrained Neural Architecture Search for Microcontrollers

Autori: Edgar Liberis, Łukasz Dudziak, Nicholas D. Lane
Pubblicato in: Proceedings of the 1st Workshop on Machine Learning and Systems (EuroMLSys), 2021
Editore: ACM

Can Fair Federated Learning reduce the need for Personalisation?

Autori: Alex Iacob, Pedro Porto Buarque Gusmão, Nicholas Lane
Pubblicato in: 2023
Editore: 3rd Workshop on Machine Learning and Systems

Protea: Client profiling within federated systems using flower

Autori: Wanru Zhao, Xinchi Qiu, Javier Fernandez-Marques, Pedro PB de Gusmão, Nicholas D Lane
Pubblicato in: 2022
Editore: 1st ACM Workshop on Data Privacy and Federated Learning Technologies for Mobile Edge Network

Deep Learning on Microcontrollers: A Studyon Deployment Costs and Challenges

Autori: Filip Svoboda Javier Fernandez-Marques Edgar Liberis Nicholas D. Lane
Pubblicato in: Proceedings of EuroMLSys, 2022
Editore: ACM
DOI: 10.1145/3517207.3526978

FedVal: Different good or different bad in federated learning

Autori: Viktor Valadi, Xinchi Qiu, Pedro Porto Buarque de Gusmão, Nicholas D. Lane, Mina Alibeigi
Pubblicato in: 2023
Editore: 32nd USENIX Security Symposium

Peer reviewed articles (2)

MobiVQA: Efficient On-Device Visual Question Answering

Autori: Qingqing Cao, Prerna Khanna, Nicholas D. Lane, Aruna Balasubramanian
Pubblicato in: ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, Numero 1, 2022, Pagina/e Volume 6 Numero 2. July 2022, ISSN 2474-9567
Editore: ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies IMWUT

A first look into the carbon footprint of federated learning

Autori: Xinchi Qiu, Titouan Parcollet, Javier Fernandez-Marques, Pedro PB de Gusmao, Yan Gao, Daniel J Beutel, Taner Topal, Akhil Mathur, Nicholas D Lane
Pubblicato in: Journal of Machine Learning Research, 2023, Pagina/e Volume 24 Numero 1. January 2023, ISSN 1533-7928
Editore: Journal of Machine Learning Research

Other (1)

Pex: Memory-efficient Microcontroller Deep Learning through Partial Execution

Autori: Edgar Liberis, Nicholas D. Lane
Pubblicato in: 2022
Editore: ArXiv pre-print