Zobrazeno 1 - 10
of 43
pro vyhledávání: '"Antun, Vegard"'
Autor:
Opsahl, Tobias A., Antun, Vegard
Most datasets used for supervised machine learning consist of a single label per data point. However, in cases where more information than just the class label is available, would it be possible to train models more efficiently? We introduce two nove
Externí odkaz:
http://arxiv.org/abs/2408.07438
Understanding the implicit regularization imposed by neural network architectures and gradient based optimization methods is a key challenge in deep learning and AI. In this work we provide sharp results for the implicit regularization imposed by the
Externí odkaz:
http://arxiv.org/abs/2307.07410
Autor:
Antun, Vegard
Recovering a signal (function) from finitely many binary or Fourier samples is one of the core problems in modern medical imaging, and by now there exist a plethora of methods for recovering a signal from such samples. Examples of methods, which can
Externí odkaz:
http://arxiv.org/abs/2106.00554
Publikováno v:
Proc. Natl. Acad. Sci. USA, 2022
Deep learning (DL) has had unprecedented success and is now entering scientific computing with full force. However, current DL methods typically suffer from instability, even when universal approximation properties guarantee the existence of stable n
Externí odkaz:
http://arxiv.org/abs/2101.08286
Methods inspired by Artificial Intelligence (AI) are starting to fundamentally change computational science and engineering through breakthrough performances on challenging problems. However, reliability and trustworthiness of such techniques is a ma
Externí odkaz:
http://arxiv.org/abs/2001.01258
There are two big unsolved mathematical questions in artificial intelligence (AI): (1) Why is deep learning so successful in classification problems and (2) why are neural nets based on deep learning at the same time universally unstable, where the i
Externí odkaz:
http://arxiv.org/abs/1906.01478
Publikováno v:
Appl. Comput. Harmon. Anal. 55 (2021) 1-40
Infinite-dimensional compressed sensing deals with the recovery of analog signals (functions) from linear measurements, often in the form of integral transforms such as the Fourier transform. This framework is well-suited to many real-world inverse p
Externí odkaz:
http://arxiv.org/abs/1905.00126
Publikováno v:
Proc. Natl. Acad. Sci. USA, 2020
Deep learning, due to its unprecedented success in tasks such as image classification, has emerged as a new tool in image reconstruction with potential to change the field. In this paper we demonstrate a crucial phenomenon: deep learning typically yi
Externí odkaz:
http://arxiv.org/abs/1902.05300
Publikováno v:
Proceedings of the National Academy of Sciences of the United States of America, 2020 Dec 01. 117(48), 30088-30095.
Externí odkaz:
https://www.jstor.org/stable/26971074
Publikováno v:
In Applied and Computational Harmonic Analysis November 2021 55:1-40