ML Anthology

Compete to Compute

Rupesh K Srivastava, Jonathan Masci, Sohrob Kazerounian, Faustino Gomez, Jürgen Schmidhuber
NeurIPS 2013 pp. 2310-2318
/neurips/2013/srivastava2013neurips-compete/

Abstract

Local competition among neighboring neurons is common in biological neural networks (NNs). We apply the concept to gradient-based, backprop-trained artificial multilayer NNs. NNs with competing linear units tend to outperform those with non-competing nonlinear units, and avoid catastrophic forgetting when training sets change over time.

PDF NeurIPS Semantic Scholar

Cite

Text

Srivastava et al. "Compete to Compute." Neural Information Processing Systems, 2013.

Markdown

[Srivastava et al. "Compete to Compute." Neural Information Processing Systems, 2013.](https://mlanthology.org/neurips/2013/srivastava2013neurips-compete/)

BibTeX

@inproceedings{srivastava2013neurips-compete,
  title     = {{Compete to Compute}},
  author    = {Srivastava, Rupesh K and Masci, Jonathan and Kazerounian, Sohrob and Gomez, Faustino and Schmidhuber, Jürgen},
  booktitle = {Neural Information Processing Systems},
  year      = {2013},
  pages     = {2310-2318},
  url       = {https://mlanthology.org/neurips/2013/srivastava2013neurips-compete/}
}