Three Decades of Activations: A Comprehensive Survey of 400 Activation Functions for Neural Networks
V Kunc, J Kléma - arxiv preprint arxiv:2402.09092, 2024 - arxiv.org
Neural networks have proven to be a highly effective tool for solving complex problems in
many areas of life. Recently, their importance and practical usability have further been …
many areas of life. Recently, their importance and practical usability have further been …
A Lipschitz spaces view of infinitely wide shallow neural networks
We revisit the mean field parametrization of shallow neural networks, using signed
measures on unbounded parameter spaces and duality pairings that take into account the …
measures on unbounded parameter spaces and duality pairings that take into account the …
Deep Networks are Reproducing Kernel Chains
Identifying an appropriate function space for deep neural networks remains a key open
question. While shallow neural networks are naturally associated with Reproducing Kernel …
question. While shallow neural networks are naturally associated with Reproducing Kernel …
Solving the Poisson Equation with Dirichlet data by shallow ReLU-networks: A regularity and approximation perspective
M Vaishampayan, S Wojtowytsch - arxiv preprint arxiv:2412.07728, 2024 - arxiv.org
For several classes of neural PDE solvers (Deep Ritz, PINNs, DeepONets), the ability to
approximate the solution or solution operator to a partial differential equation (PDE) hinges …
approximate the solution or solution operator to a partial differential equation (PDE) hinges …
Learning a Sparse Representation of Barron Functions with the Inverse Scale Space Flow
This paper presents a method for finding a sparse representation of Barron functions.
Specifically, given an $ L^ 2$ function $ f $, the inverse scale space flow is used to find a …
Specifically, given an $ L^ 2$ function $ f $, the inverse scale space flow is used to find a …
Inference genové exprese pomocí umělých neuronových sítí
K Vladimír - 2024 - dspace.cvut.cz
Merení genové exprese je nezbytné pro porozumení bunecným procesum a stavum v
rozlicných experimentálních podmínkách, což je potreba v ruzných oblastech …
rozlicných experimentálních podmínkách, což je potreba v ruzných oblastech …
Inference genové exprese pomocí umělých neuronových sítí
V Kunc - 2024 - search.proquest.com
Merení genové exprese je nezbytné pro porozumení bunecným procesum a stavum v
rozlicných experimentálních podmínkách, což je potreba v ruzných oblastech …
rozlicných experimentálních podmínkách, což je potreba v ruzných oblastech …
Shallow Neural Networks and Laplace's Equation on the Half-Space with Dirichlet Boundary Data
M Vaishampayan - Pittsburgh Interdisciplinary Mathematics Review, 2024 - pimr.pitt.edu
In this paper we investigate the ability of Shallow Neural Networks ie neural networks with
one hidden layer, to solve Laplace's equation on the half space. We are interested in …
one hidden layer, to solve Laplace's equation on the half space. We are interested in …