Zobrazeno 1 - 10
of 1 293
pro vyhledávání: '"A. Ben-Shaul"'
Self-supervised learning (SSL) is a powerful tool in machine learning, but understanding the learned representations and their underlying mechanisms remains a challenge. This paper presents an in-depth empirical analysis of SSL-trained representation
Externí odkaz:
http://arxiv.org/abs/2305.15614
Multiplication layers are a key component in various influential neural network modules, including self-attention and hypernetwork layers. In this paper, we investigate the approximation capabilities of deep neural networks with intermediate neurons
Externí odkaz:
http://arxiv.org/abs/2301.04605
Autor:
Rokni, Dan, Ben-Shaul, Yoram
Publikováno v:
In Trends in Neurosciences October 2024 47(10):834-848
Recent results in the literature suggest that the penultimate (second-to-last) layer representations of neural networks that are trained for classification exhibit a clustering property called neural collapse (NC). We study the implicit bias of stoch
Externí odkaz:
http://arxiv.org/abs/2202.09028
Autor:
Ben-Shaul, Ido, Dekel, Shai
Recent advances in theoretical Deep Learning have introduced geometric properties that occur during training, past the Interpolation Threshold -- where the training error reaches zero. We inquire into the phenomena coined Neural Collapse in the inter
Externí odkaz:
http://arxiv.org/abs/2201.08924
Autor:
Maximilian Nagel, Marco Niestroj, Rohini Bansal, David Fleck, Angelika Lampert, Romana Stopkova, Pavel Stopka, Yoram Ben-Shaul, Marc Spehr
Publikováno v:
eLife, Vol 12 (2024)
In most mammals, conspecific chemosensory communication relies on semiochemical release within complex bodily secretions and subsequent stimulus detection by the vomeronasal organ (VNO). Urine, a rich source of ethologically relevant chemosignals, co
Externí odkaz:
https://doaj.org/article/22c1aa4d29f14708bc6cc931ab44a5fa
Autor:
Hamacher, Christoph, Degen, Rudolf, Franke, Melissa, Switacz, Victoria K., Fleck, David, Katreddi, Raghu Ram, Hernandez-Clavijo, Andres, Strauch, Martin, Horio, Nao, Hachgenei, Enno, Spehr, Jennifer, Liberles, Stephen D., Merhof, Dorit, Forni, Paolo E., Zimmer-Bensch, Geraldine, Ben-Shaul, Yoram, Spehr, Marc
Publikováno v:
In Current Biology 25 March 2024 34(6):1206-1221
Autor:
Ben-Shaul, Ido, Dekel, Shai
We propose a probe for the analysis of deep learning architectures that is based on machine learning and approximation theoretical principles. Given a deep learning architecture and a training set, during or after training, the Sparsity Probe allows
Externí odkaz:
http://arxiv.org/abs/2105.06849
Multiple Instance Learning is a form of weakly supervised learning in which the data is arranged in sets of instances called bags with one label assigned per bag. The bag level class prediction is derived from the multiple instances through applicati
Externí odkaz:
http://arxiv.org/abs/2008.10548
In this work, we explore the ability of NN (Neural Networks) to serve as a tool for finding eigen-pairs of ordinary differential equations. The question we aime to address is whether, given a self-adjoint operator, we can learn what are the eigenfunc
Externí odkaz:
http://arxiv.org/abs/2007.10205