Mitigating severe over-parameterization in deep convolutional neural networks through forced feature abstraction and compression with an entropy-based heuristic


by Gowdra, N, Sinha, R, MacDonell, S and Yan, WQ
Reference:
Mitigating severe over-parameterization in deep convolutional neural networks through forced feature abstraction and compression with an entropy-based heuristic (Gowdra, N, Sinha, R, MacDonell, S and Yan, WQ), In Pattern Recognition, Elsevier BV, 2021.
Bibtex Entry:
@article{gowdra2021mitigatingheuristic,
author = {Gowdra, N and Sinha, R and MacDonell, S and Yan, WQ},
journal = {Pattern Recognition},
number = {108057},
publisher = {Elsevier BV},
title = {Mitigating severe over-parameterization in deep convolutional neural networks through forced feature abstraction and compression with an entropy-based heuristic},
year = {2021},
doi = {10.1016/j.patcog.2021.108057},
issn = {0031-3203},
language = {en},
}