Stationary points of a shallow neural network with quadratic activations and the global optimality of the gradient descent algorithm
Year of publication: |
2025
|
---|---|
Authors: | Gamarnik, David ; Kızıldağ, Eren C. ; Zadik, Ilias |
Published in: |
Mathematics of operations research. - Hanover, Md. : INFORMS, ISSN 1526-5471, ZDB-ID 2004273-5. - Vol. 50.2025, 1, p. 209-251
|
Subject: | empirical risk minimization | generalization | gradient descent | initialization | neural networks | optimization landscape | semicircle law | Neuronale Netze | Neural networks | Mathematische Optimierung | Mathematical programming | Theorie | Theory | Algorithmus | Algorithm |
-
An evaluation of the bihyperbolic function in the optimization of the backpropagation algorithm
Miguez, Geraldo, (2014)
-
Scalability of using restricted Boltzmann machines for combinatorial optimization
Probst, Malte, (2017)
-
An optimised neural network-based spectrum prediction scheme for cognitive radio
Bhuvaneswari, B., (2020)
- More ...
-
Mixed-integer convex representability
Lubin, Miles, (2022)
-
An improved upper bound for the TSP in cubic 3-edge-connected graphs
Gamarnik, David, (2005)
-
Bertsimas, Dimitris, (2003)
- More ...