Learning and Optimization in Luminy - LOL2022 / Apprentissage et Optimisation à Luminy - LOL2022

Collection Learning and Optimization in Luminy - LOL2022 / Apprentissage et Optimisation à Luminy - LOL2022

Organisateur(s) Boyer, Claire ; d'Aspremont, Alexandre ; Dieuleveut, Aymeric ; Moreau, Thomas ; Villar, Soledad

Date(s) 03/10/2022 - 07/10/2022

URL associée https://conferences.cirm-math.fr/2551.html

00:00:00 / 00:00:00

1 5

Private frequency estimation via projective geometry

De Jelani Nelson

Many of us use smartphones and rely on tools like auto-complete and spelling auto-correct to make using these devices more pleasant, but building these tools presents a challenge. On the one hand, the machine-learning algorithms used to provide these features require data to learn from, but on the other hand, who among us is willing to send a carbon copy of all our text messages to device manufacturers to provide that data? 'Local differential privacy' and related concepts have emerged as the gold standard model in which to analyze tradeoffs between losses in utility and privacy for solutions to such problems. In this talk, we give a new state-of-the-art algorithm for estimating histograms of user data, making use of projective geometry over finite fields coupled with a reconstruction algorithm based on dynamic programming. This talk is based on joint work with Vitaly Feldman (Apple), Huy Le Nguyen (Northeastern), and Kunal Talwar (Apple).

Informations sur la vidéo

Date de captation 03/10/2022
Date de publication 10/11/2022
Institut CIRM
Licence CC BY NC ND
Langue Anglais
Audience Chercheurs, Doctorants
Réalisateur(s) Guillaume Hennenfent
Format MP4

Données de citation

DOI 10.24350/CIRM.V.19965703
Citer cette vidéo Nelson, Jelani (03/10/2022). Private frequency estimation via projective geometry. CIRM. Audiovisual resource. DOI: 10.24350/CIRM.V.19965703
URL https://dx.doi.org/10.24350/CIRM.V.19965703

Domaine(s)

Cryptographie et sécurité

Bibliographie

FELDMAN, Vitaly, NELSON, Jelani, NGUYEN, Huy, et al. Private frequency estimation via projective geometry. In : International Conference on Machine Learning. PMLR, 2022. p. 6418-6433. - https://proceedings.mlr.press/v162/feldman22a.html

Codes MSC

Document(s)

https://www.cirm-math.fr/RepOrga/2551/Slides/nelson_compressed.pdf

Dernières questions liées sur MathOverflow

Pour poser une question, votre compte Carmin.tv doit être connecté à mathoverflow

Poser une question sur MathOverflow

Toutes les vidéos de la collection

48:34

publiée le 10 novembre 2022

Private frequency estimation via projective geometry

De Jelani Nelson

35:00

publiée le 10 novembre 2022

Generalisation of some overparametrised models

De Stéphane Chrétien

35:47

publiée le 10 novembre 2022

Curiosities and counterexamples in smooth convex optimization

De Edouard Pauwels

47:22

publiée le 10 novembre 2022

Non-convex SGD and Lojasiewicz-type conditions for deep learning

De Kevin Scaman

42:41

publiée le 10 novembre 2022

Stochastic normalizing flows and the power of patches in inverse problems

De Gabriele Steidl

Copyright Carmin.tv 2024

Donner son avis