Zobrazeno 1 - 10
of 4 311
pro vyhledávání: '"KORNOWSKI, A."'
We study differentially private (DP) optimization algorithms for stochastic and empirical objectives which are neither smooth nor convex, and propose methods that return a Goldstein-stationary point with sample complexity bounds that improve on exist
Externí odkaz:
http://arxiv.org/abs/2410.05880
Autor:
Kornowski, Guy
We present differentially private (DP) algorithms for bilevel optimization, a problem class that received significant attention lately in various machine learning applications. These are the first DP algorithms for this task that are able to provide
Externí odkaz:
http://arxiv.org/abs/2409.19800
We study the oracle complexity of nonsmooth nonconvex optimization, with the algorithm assumed to have access only to local function information. It has been shown by Davis, Drusvyatskiy, and Jiang (2023) that for nonsmooth Lipschitz functions satisf
Externí odkaz:
http://arxiv.org/abs/2409.10323
Autor:
Kornowski, Guy, Shamir, Ohad
Recent results show that vanilla gradient descent can be accelerated for smooth convex objectives, merely by changing the stepsize sequence. We show that this can lead to surprisingly large errors indefinitely, and therefore ask: Is there any stepsiz
Externí odkaz:
http://arxiv.org/abs/2406.13888
Algorithms for bilevel optimization often encounter Hessian computations, which are prohibitive in high dimensions. While recent works offer first-order methods for unconstrained bilevel problems, the constrained setting remains relatively underexplo
Externí odkaz:
http://arxiv.org/abs/2406.12771
We study distribution-free nonparametric regression following a notion of average smoothness initiated by Ashlagi et al. (2021), which measures the "effective" smoothness of a function with respect to an arbitrary unknown underlying distribution. Whi
Externí odkaz:
http://arxiv.org/abs/2309.17016
Publikováno v:
Cardiovascular Diabetology, Vol 23, Iss 1, Pp 1-12 (2024)
Abstract Background Diabetes mellitus (DM) is associated with premature atherosclerotic disease, coronary artery disease (CAD) and chronic heart failure (HF), leading to increased morbidity and mortality. Sodium-Glucose Co-transporter 2 Inhibitors (S
Externí odkaz:
https://doaj.org/article/69d2b3753a9c4d0cac705f66ab828976
Autor:
Kornowski, Guy, Shamir, Ohad
We study the complexity of producing $(\delta,\epsilon)$-stationary points of Lipschitz objectives which are possibly neither smooth nor convex, using only noisy function evaluations. Recent works proposed several stochastic zero-order algorithms tha
Externí odkaz:
http://arxiv.org/abs/2307.04504
Overparameterized neural networks (NNs) are observed to generalize well even when trained to perfectly fit noisy data. This phenomenon motivated a large body of work on "benign overfitting", where interpolating predictors achieve near-optimal perform
Externí odkaz:
http://arxiv.org/abs/2305.15141
Autor:
Idan Goldberg, Uri Landes, Genady Drozdinsky, Pablo Codner, Tamir Bental, Katia Orvin, Nili Schamroth Pravda, Lotem Goldberg, Omri Soudry, Tsahi T. Lerman, Ran Kornowski, Alon Eisen, Hana Vaknin-Assa
Publikováno v:
Pulse, Vol 12, Iss 1, Pp 85-94 (2024)
Introduction: Transcatheter aortic valve implantation (TAVI) has emerged as a common and effective treatment for patients with severe aortic stenosis. Changes in systemic blood pressure after TAVI have been described, yet their prognostic value is no
Externí odkaz:
https://doaj.org/article/66464b15e02f45f5a02704620d946ff4