Zobrazeno 1 - 10
of 6 085
pro vyhledávání: '"Kunin IS"'
Recent breakthroughs in high repetition-rate extreme ultraviolet (XUV) light sources and photoelectron analyzers have led to dramatic advances in the performance of time-resolved photoemission experiments. In this perspective article, we discuss the
Externí odkaz:
http://arxiv.org/abs/2410.11589
Autor:
Dominé, Clémentine C. J., Anguita, Nicolas, Proca, Alexandra M., Braun, Lukas, Kunin, Daniel, Mediano, Pedro A. M., Saxe, Andrew M.
Biological and artificial neural networks develop internal representations that enable them to perform complex tasks. In artificial networks, the effectiveness of these models relies on their ability to build task specific representation, a process i
Externí odkaz:
http://arxiv.org/abs/2409.14623
Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning
Autor:
Kunin, Daniel, Raventós, Allan, Dominé, Clémentine, Chen, Feng, Klindt, David, Saxe, Andrew, Ganguli, Surya
While the impressive performance of modern neural networks is often attributed to their capacity to efficiently extract task-relevant features from data, the mechanisms underlying this rich feature learning regime remain elusive, with much of our the
Externí odkaz:
http://arxiv.org/abs/2406.06158
Autor:
Bakalis, Jin, Chernov, Sergii, Li, Ziling, Kunin, Alice, Withers, Zachary H., Cheng, Shuyu, Adler, Alexander, Zhao, Peng, Corder, Christopher, White, Michael G., Schönhense, Gerd, Du, Xu, Kawkami, Roland, Allison, Thomas K.
Publikováno v:
Nano Lett. 24, 9353 (2024)
The unique optical properties of graphene, with broadband absorption and ultrafast response, make it a critical component of optoelectronic and spintronic devices. Using time-resolved momentum microscopy with high data rate and high dynamic range, we
Externí odkaz:
http://arxiv.org/abs/2402.13205
In this work, we reveal a strong implicit bias of stochastic gradient descent (SGD) that drives overly expressive networks to much simpler subnetworks, thereby dramatically reducing the number of independent parameters, and improving generalization.
Externí odkaz:
http://arxiv.org/abs/2306.04251
Autor:
Christopher P. Carr, Allan D. Tate, Amanda Trofholz, Junia N. de Brito, Andrea N. Trejo, Michael F. Troy, Jerica M. Berge, Alicia Kunin-Batson
Publikováno v:
Health Equity, Vol 8, Iss 1, Pp 355-359 (2024)
Introduction: Past research shows that structural racism contributes to disparities in cardiometabolic health among racially/ethnically minoritized populations. Methods: This cross-sectional study examined the correlation between census tract-level r
Externí odkaz:
https://doaj.org/article/3f0015e05d984cf4a73acb3d99456290
Autor:
Alicia S. Kunin‐Batson, A. Lauren Crain, Nancy E. Sherwood, Aaron S. Kelly, Elyse O. Kharbanda, Megan R. Gunnar, Jacob Haapala, Elisabeth M. Seburg, Simone A. French
Publikováno v:
Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease, Vol 13, Iss 18 (2024)
Background Previous studies have found that exposure to childhood environmental stress is associated with cardiometabolic risk. However, it is not known whether individual health behaviors disrupt this relationship. This study prospectively evaluated
Externí odkaz:
https://doaj.org/article/1297267aa2a942d09cf92d82973b0ac5
In this work, we explore the maximum-margin bias of quasi-homogeneous neural networks trained with gradient flow on an exponential loss and past a point of separability. We introduce the class of quasi-homogeneous models, which is expressive enough t
Externí odkaz:
http://arxiv.org/abs/2210.03820
In this article, we propose the approach to structural optimization of neural networks, based on the braid theory. The paper describes the basics of braid theory as applied to the description of graph structures of neural networks. It is shown how ne
Externí odkaz:
http://arxiv.org/abs/2207.04121
Publikováno v:
Interface: Comunicação, Saúde, Educação, Vol 28 (2024)
Externí odkaz:
https://doaj.org/article/d322554ca9b44d378eee7a81afad3257