Grid-Functioned Neural Networks

Javier Dehesa, Andrew Vidler, Julian Padget, Christof Lutteroth
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:2559-2567, 2021.

Abstract

We introduce a new neural network architecture that we call "grid-functioned" neural networks. It utilises a grid structure of network parameterisations that can be specialised for different subdomains of the problem, while maintaining smooth, continuous behaviour. The grid gives the user flexibility to prevent gross features from overshadowing important minor ones. We present a full characterisation of its computational and spatial complexity, and demonstrate its potential, compared to a traditional architecture, over a set of synthetic regression problems. We further illustrate the benefits through a real-world 3D skeletal animation case study, where it offers the same visual quality as a state-of-the-art model, but with lower computational complexity and better control accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-dehesa21a, title = {Grid-Functioned Neural Networks}, author = {Dehesa, Javier and Vidler, Andrew and Padget, Julian and Lutteroth, Christof}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {2559--2567}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/dehesa21a/dehesa21a.pdf}, url = {https://proceedings.mlr.press/v139/dehesa21a.html}, abstract = {We introduce a new neural network architecture that we call "grid-functioned" neural networks. It utilises a grid structure of network parameterisations that can be specialised for different subdomains of the problem, while maintaining smooth, continuous behaviour. The grid gives the user flexibility to prevent gross features from overshadowing important minor ones. We present a full characterisation of its computational and spatial complexity, and demonstrate its potential, compared to a traditional architecture, over a set of synthetic regression problems. We further illustrate the benefits through a real-world 3D skeletal animation case study, where it offers the same visual quality as a state-of-the-art model, but with lower computational complexity and better control accuracy.} }
Endnote
%0 Conference Paper %T Grid-Functioned Neural Networks %A Javier Dehesa %A Andrew Vidler %A Julian Padget %A Christof Lutteroth %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-dehesa21a %I PMLR %P 2559--2567 %U https://proceedings.mlr.press/v139/dehesa21a.html %V 139 %X We introduce a new neural network architecture that we call "grid-functioned" neural networks. It utilises a grid structure of network parameterisations that can be specialised for different subdomains of the problem, while maintaining smooth, continuous behaviour. The grid gives the user flexibility to prevent gross features from overshadowing important minor ones. We present a full characterisation of its computational and spatial complexity, and demonstrate its potential, compared to a traditional architecture, over a set of synthetic regression problems. We further illustrate the benefits through a real-world 3D skeletal animation case study, where it offers the same visual quality as a state-of-the-art model, but with lower computational complexity and better control accuracy.
APA
Dehesa, J., Vidler, A., Padget, J. & Lutteroth, C.. (2021). Grid-Functioned Neural Networks. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:2559-2567 Available from https://proceedings.mlr.press/v139/dehesa21a.html.

Related Material