Information Theory beyond Communications: Distributed Representations and Deep Learning

Informazioni relative al progetto

STRUDEL

ID dell’accordo di sovvenzione: 797805

DOI

10.3030/797805

Progetto chiuso

Data della firma CE 5 Aprile 2018

Data di avvio 1 Settembre 2018

Data di completamento 31 Agosto 2020

Finanziato da

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Costo totale

€ 171 349,20

Contributo UE

€ 171 349,20

Coordinato da

CENTRALESUPELEC
France

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 (13)

Detecting Covariate Shift with Black Box Predictors

Autori: Florence Alberge, Clement Feutry, Pierre Duhamel, Pablo Piantanida
Pubblicato in: 2019 26th International Conference on Telecommunications (ICT), 2019, Pagina/e 324-329, ISBN 978-1-7281-0273-3
Editore: IEEE
DOI: 10.1109/ict.2019.8798827

Deep Directed Information-Based Learning for Privacy-Preserving Smart Meter Data Release

Autori: Mohammadhadi Shateri, Francisco Messina, Pablo Piantanida, Fabrice Labeau
Pubblicato in: 2019 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2019, Pagina/e 1-7, ISBN 978-1-5386-8099-5
Editore: IEEE
DOI: 10.1109/smartgridcomm.2019.8909813

Privacy-Preserving Synthetic Smart Meters Data

Autori: Ganesh Del Grosso, Georg Pichler, Pablo Piantanida
Pubblicato in: 2021 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), 2021, Pagina/e 1-5, ISBN 978-1-7281-8897-3
Editore: IEEE
DOI: 10.1109/isgt49243.2021.9372157

Few-Shot Segmentation Without Meta-Learning: A Good Transductive Inference Is All You Need?

Autori: Malik Boudiaf, Hoel Kervadec, Ziko Imtiaz Masud, Pablo Piantanida, Ismail Ben Ayed, Jose Dolz
Pubblicato in: IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, Pagina/e 13979-13988, ISBN 978-1-6654-4509-2
Editore: IEEE
DOI: 10.1109/cvpr46437.2021.01376

A Differential Entropy Estimator for Training Neural Networks

Autori: Georg Pichler, Pierre Jean A. Colombo, Malik Boudiaf, Günther Koliander, Pablo Piantanida
Pubblicato in: International Conference on Machine Learning, Numero 162, 2022, Pagina/e 17691-17715
Editore: PMLR

DOCTOR: A Simple Method for Detecting Misclassification Errors

Autori: Federica Granese, Marco Romanelli, Daniele Gorla, Catuscia Palamidessi, Pablo Piantanida
Pubblicato in: Advances in Neural Information Processing Systems, Numero 34, 2021, Pagina/e 5669-5681, ISSN 1049-5258
Editore: Curran Associates, Inc.
DOI: 10.48550/arxiv.2106.02395

A Novel Estimator of Mutual Information for Learning to Disentangle Textual Representations

Autori: Pierre Colombo, Pablo Piantanida, Chloé Clavel
Pubblicato in: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), 2021, Pagina/e 6539–6550
Editore: Association for Computational Linguistics
DOI: 10.18653/v1/2021.acl-long.511

Transductive Information Maximization For Few-Shot Learning

Autori: Malik Boudiaf, Ziko Imtiaz Masud, Jérôme Rony, Jose Dolz, Pablo Piantanida,Ismail Ben Ayed
Pubblicato in: Conference on Neural Information Processing Systems, Numero 33, 2020, ISBN 9781713829546
Editore: Curran Associates, Inc.

Privacy-Cost Management in Smart Meters Using Deep Reinforcement Learning

Autori: Mohammadhadi Shateri, Francisco Messina, Pablo Piantanida, Fabrice Labeau
Pubblicato in: 2020 IEEE PES Innovative Smart Grid Technologies Europe (ISGT-Europe), 2020, Pagina/e 929-933, ISBN 978-1-7281-7100-5
Editore: IEEE
DOI: 10.1109/isgt-europe47291.2020.9248831

Estimating g-leakage via machine learning

Autori: K. Chatzikokolakis, C. Palamidessi, P. Piantanida, and M. Romanelli
Pubblicato in: ACM Conference on Computer and Communications Security (CCS), 2020
Editore: ACM
DOI: 10.1145/3372297.3423363

On the Impact of Side Information on Smart Meter Privacy-Preserving Methods

Autori: Mohammadhadi Shateri, Francisco Messina, Pablo Piantanida, Fabrice Labeau
Pubblicato in: 2020 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm), 2020, Pagina/e 1-6, ISBN 978-1-7281-6127-3
Editore: IEEE
DOI: 10.1109/smartgridcomm47815.2020.9302948

Learning Semi-Supervised Anonymized Representations by Mutual Information

Autori: C. Feutry, P. Piantanida, P. Duhamel
Pubblicato in: ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020, Pagina/e 3467-3471, ISBN 978-1-5090-6631-5
Editore: IEEE
DOI: 10.1109/icassp40776.2020.9053379

On Direct Distribution Matching for Adapting Segmentation Networks SAVE

Autori: Pichler, Georg; Dolz, Jose; Ayed, Ismail Ben; Piantanida, Pablo
Pubblicato in: MIDL 2020 - Medical Imaging with Deep Learning Publications Manually add publicatio, Numero 3, 2020
Editore: PMLR
DOI: 10.48550/arXiv.1904.02657

Other (2)

Learning anonymized representations with adversarial neural networks

Autori: C. Feutry, P. Piantanida, Y. Bengio, and P. Duhamel
Pubblicato in: Technical report, 2018
Editore: Online technical report

On the Estimation of Information Measures of Continuous Distributions

Autori: Georg Pichler, Pablo Piantanida, Günther Koliander
Pubblicato in: Technical report, 2020
Editore: Technical Report

Peer reviewed articles (10)

On Universal D-Semifaithful Coding for Memoryless Sources With Infinite Alphabets

Autori: Jorge F. Silva, Pablo Piantanida
Pubblicato in: IEEE Transactions on Information Theory, Numero Volume: 68, Numero: 4, 2022, Pagina/e 1557-9654, ISSN 0018-9448
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tit.2021.3134891

Finite-Length Bounds on Hypothesis Testing Subject to Vanishing Type I Error Restrictions

Autori: Sebastian Espinosa, Jorge F. Silva, Pablo Piantanida
Pubblicato in: IEEE Signal Processing Letters, Numero 28, 2021, Pagina/e 229-233, ISSN 1070-9908
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/lsp.2021.3050381

Collaborative Information Bottleneck

Autori: Matias Vera, Leonardo Rey Vega, Pablo Piantanida
Pubblicato in: IEEE Transactions on Information Theory, Numero 65/2, 2019, Pagina/e 787-815, ISSN 0018-9448
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tit.2018.2883295

The Role of Mutual Information in Variational Classifiers

Autori: Matias Vera, Leonardo Rey Vega, Pablo Piantanida
Pubblicato in: Under revision for publication in Machine Learning, 2023, ISSN 1532-4435
Editore: MIT Press
DOI: 10.1007/s10994-023-06337-6

Distributed information-theoretic clustering

Autori: Georg Pichler, Pablo Piantanida, Gerald Matz
Pubblicato in: Information and Inference: A Journal of the IMA, 2021, ISSN 2049-8772
Editore: Oxford University Press
DOI: 10.1093/imaiai/iaab007

Fundamental Limits of Decentralized Data Shuffling

Autori: Kai Wan, Daniela Tuninetti, Mingyue Ji, Giuseppe Caire, Pablo Piantanida
Pubblicato in: IEEE Transactions on Information Theory, Numero 66/6, 2020, Pagina/e 3616-3637, ISSN 0018-9448
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tit.2020.2966197

On the Exponential Approximation of Type II Error Probability of Distributed Test of Independence

Autori: Espinosa, Sebastian; Silva, Jorge F.; Piantanida, Pablo
Pubblicato in: IEEE Transactions on Signal and Information Processing over Networks, 2021, Pagina/e 777 - 790, ISSN 2373-776X
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tsipn.2021.3133192

Adversarial Robustness Via Fisher-Rao Regularization

Autori: Marine Picot, Francisco Messina, Malik Boudiaf, Fabrice Labeau, Ismail Ben Ayed, Pablo Piantanida
Pubblicato in: IEEE Transactions on Pattern Analysis and Machine Intelligence, Numero vol. 45, no. 3, 2023, Pagina/e 2698-2710, ISSN 0162-8828
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tpami.2022.3174724

Privacy-Cost Management in Smart Meters with Mutual Information-Based Reinforcement Learning

Autori: M. Shateri, F. Messina, P. Piantanida, and F. Labeau
Pubblicato in: IEEE Internet of Things Journal, 2021, Pagina/e 1 - 1, ISSN 2327-4662
Editore: Institute of Electrical and Electronics Engineers Inc.
DOI: 10.1109/jiot.2021.3128488

Real-Time Privacy-Preserving Data Release for Smart Meters

Autori: Mohammadhadi Shateri, Francisco Messina, Pablo Piantanida, Fabrice Labeau
Pubblicato in: IEEE Transactions on Smart Grid, 2020, Pagina/e 1-1, ISSN 1949-3053
Editore: Institute of Electrical and Electronics Engineers
DOI: 10.1109/tsg.2020.3005634

Book chapters (2)

Information Bottleneck and Representation Learning

Autori: Pablo Piantanida, Leonardo Rey Vega
Pubblicato in: Information-Theoretic Methods in Data Science, 2021, Pagina/e 330-358, ISBN 9781108616799
Editore: Cambridge University Press
DOI: 10.1017/9781108616799.012

A Unifying Mutual Information View of Metric Learning: Cross-Entropy vs. Pairwise Losses

Autori: Malik Boudiaf, Jérôme Rony, Imtiaz Masud Ziko, Eric Granger, Marco Pedersoli, Pablo Piantanida, Ismail Ben Ayed
Pubblicato in: Computer Vision – ECCV 2020 - 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VI, Numero 12351, 2020, Pagina/e 548-564, ISBN 978-3-030-58538-9
Editore: Springer International Publishing
DOI: 10.1007/978-3-030-58539-6_33