Zobrazeno 1 - 10
of 12
pro vyhledávání: '"Giovanni Cherubin"'
Publikováno v:
Proceedings on Privacy Enhancing Technologies, Vol 2022, Iss 1, Pp 460-480 (2022)
A membership inference attack (MIA) against a machine-learning model enables an attacker to determine whether a given data record was part of the model's training data or not. In this paper, we provide an in-depth study of the phenomenon of disparate
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fee72e31fcb0bc9b57846b925f83f812
https://doaj.org/article/c3bf6102ce1b4060b1996827dc5cbec5
https://doaj.org/article/c3bf6102ce1b4060b1996827dc5cbec5
Autor:
Ana Stanojevic, Stanisław Woźniak, Guillaume Bellec, Giovanni Cherubini, Angeliki Pantazi, Wulfram Gerstner
Publikováno v:
Nature Communications, Vol 15, Iss 1, Pp 1-13 (2024)
Abstract Communication by rare, binary spikes is a key factor for the energy efficiency of biological brains. However, it is harder to train biologically-inspired spiking neural networks than artificial neural networks. This is puzzling given that th
Externí odkaz:
https://doaj.org/article/232a98336e0941d79563834bb5c1b24e
Autor:
Giovanni Cherubin
Publikováno v:
Machine Learning. 108:475-488
We study majority vote ensembles of $$\varepsilon $$ -valid conformal predictors (CP). We show that the prediction set $$\varGamma ^\eta $$ produced as the majority vote among the prediction sets $$\varGamma ^\varepsilon _i$$ of k independent $$\vare
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::de32a22a7d649f3f808cf63b9f558bca
https://doi.org/10.1007/s10994-018-5752-y
https://doi.org/10.1007/s10994-018-5752-y
Autor:
Stanisław Woźniak, Hlynur Jónsson, Giovanni Cherubini, Angeliki Pantazi, Evangelos Eleftheriou
Publikováno v:
Nature Communications, Vol 14, Iss 1, Pp 1-9 (2023)
Abstract Visual oddity task was conceived to study universal ethnic-independent analytic intelligence of humans from a perspective of comprehension of spatial concepts. Advancements in artificial intelligence led to important breakthroughs, yet excel
Externí odkaz:
https://doaj.org/article/bf8b556814a54eb58aa4d8ad8b929268
Autor:
Giovanni Cherubin
Publikováno v:
Proceedings on Privacy Enhancing Technologies, Vol 2017, Iss 4, Pp 215-231 (2017)
Website Fingerprinting (WF) attacks raise major concerns about users’ privacy. They employ Machine Learning (ML) techniques to allow a local passive adversary to uncover the Web browsing behavior of a user, even if she browses through an encrypted
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::a011f7639a30582f93ada769b709ff96
https://doaj.org/article/b05e98d4fb2b4b0ba483137de65ec4a4
https://doaj.org/article/b05e98d4fb2b4b0ba483137de65ec4a4
Publikováno v:
Proceedings on Privacy Enhancing Technologies, Vol 2017, Iss 2, Pp 186-203 (2017)
Proceedings on Privacy Enhancing Technologies
Cherubin, G, Hayes, J & Juarez, M 2017, ' Website Fingerprinting Defenses at the Application Layer ', Proceedings on Privacy Enhancing Technologies, vol. 2017, no. 2, pp. 168-185 . https://doi.org/10.1515/popets-2017-0023
Proceedings on Privacy Enhancing Technologies
Cherubin, G, Hayes, J & Juarez, M 2017, ' Website Fingerprinting Defenses at the Application Layer ', Proceedings on Privacy Enhancing Technologies, vol. 2017, no. 2, pp. 168-185 . https://doi.org/10.1515/popets-2017-0023
Website Fingerprinting (WF) allows a passive network adversary to learn the websites that a client visits by analyzing traffic patterns that are unique to each website. It has been recently shown that these attacks are particularly effective against
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::52267168bc16c09b6b1e322824107444
https://doaj.org/article/595c1a27fb764346981cfca8441eac87
https://doaj.org/article/595c1a27fb764346981cfca8441eac87
Autor:
Giovanni Cherubin, Ilia Nouretdinov
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783319333946
COPA
COPA
We consider the problem of training a Hidden Markov Model HMM from fully observable data and predicting the hidden states of an observed sequence. Our attention is focused to applications that require a list of potential sequences as a prediction. We
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::172c14ac06d34f587556b401be733098
https://doi.org/10.1007/978-3-319-33395-3_10
https://doi.org/10.1007/978-3-319-33395-3_10
Autor:
Zhi Wang, Alexander Gammerman, Davide Papini, Lorenzo Cavallaro, Giovanni Cherubin, Roberto Jordaney, Ilia Nouretdinov
Publikováno v:
Statistical Learning and Data Sciences ISBN: 9783319170909
SLDS
SLDS
The paper describes an application of a novel clustering technique based on Conformal Predictors. Unlike traditional clustering methods, this technique allows to control the number of objects that are left outside of any cluster by setting up a requi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::f600f34710b6de5774e5bf38cf167993
https://doi.org/10.1007/978-3-319-17091-6_26
https://doi.org/10.1007/978-3-319-17091-6_26
Autor:
Geethan Karunaratne, Manuel Schmuck, Manuel Le Gallo, Giovanni Cherubini, Luca Benini, Abu Sebastian, Abbas Rahimi
Publikováno v:
Nature Communications, Vol 12, Iss 1, Pp 1-12 (2021)
The implementation of memory-augmented neural networks using conventional computer architectures is challenging due to a large number of read and write operations. Here, Karunaratne, Schmuck et al. propose an architecture that enables analog in-memor
Externí odkaz:
https://doaj.org/article/b553751b28d34003af781f5a72f0719a
Publikováno v:
Entropy, Vol 22, Iss 7, p 727 (2020)
Information theory concepts are leveraged with the goal of better understanding and improving Deep Neural Networks (DNNs). The information plane of neural networks describes the behavior during training of the mutual information at various depths bet
Externí odkaz:
https://doaj.org/article/be44bb0302404f4ca8555930ba67fa97