Editors
3
Impact
Loading...
Mean results of algorithms applied to SGL1 for Lenet-5 models trained on MNIST with lowest test errors across 5 runs when varying the regularization parameter λ=α/60000 when α∈{0.1,0.2,0.3,0.4,0.5}. (A) Mean test error. (B) Mean weight sparsity. (C) Mean neuron sparsity.
5,344 views
12 citations
3,173 views
12 citations
3,281 views
10 citations
Recommended Research Topics
Frontiers Logo

Frontiers in Applied Mathematics and Statistics

Compressed Sensing over Complex Networks for Learning from Big Data over Networks
Edited by Alex Jung, Holger Rauhut, Sergiy A. Vorobyov
25K
views
4
authors
4
articles
Frontiers Logo

Frontiers in Applied Mathematics and Statistics

Computational Methods for Data Analytics
Edited by Stephane Chretien, Christophe Biernacki
17.4K
views
19
authors
4
articles
Frontiers Logo

Frontiers in Applied Mathematics and Statistics

Mathematical Fundamentals of Machine Learning
Edited by David Glickenstein, Keaton Hamm, Xiaoming Huo, Yajun Mei, Martin Stoll
21.8K
views
17
authors
6
articles
Frontiers Logo

Frontiers in Applied Mathematics and Statistics

High-Performance Tensor Computations in Scientific Computing and Data Science
Edited by Edoardo Angelo Di Napoli, Paolo Bientinesi, Jiajia Li, André Uschmajew
52.8K
views
43
authors
11
articles
Frontiers Logo

Frontiers in Applied Mathematics and Statistics

The Interplay of Discrete Optimization and Machine Learning
Edited by Andreas M. Tillmann, Gonzalo Muñoz, Sebastian Pokutta, Elias Khalil
17.8K
views
8
authors
3
articles