Competition and Multiple Cause Models

Abstract

If different causes can interact on any occasion to generate a set of patterns, then systems modeling the generation have to model the interaction too. We discuss a way of combining multiple causes that is based on the Integrated Segmentation and Recognition architecture of Keeler et al. (1991). It is more cooperative than the scheme embodied in the mixture of experts architecture, which insists that just one cause generate each output, and more competitive than the noisy-or combination function, which was recently suggested by Saund (1994a, b). Simulations confirm its efficacy.

Cite

Text

Dayan and Zemel. "Competition and Multiple Cause Models." Neural Computation, 1995. doi:10.1162/NECO.1995.7.3.565

Markdown

[Dayan and Zemel. "Competition and Multiple Cause Models." Neural Computation, 1995.](https://mlanthology.org/neco/1995/dayan1995neco-competition/) doi:10.1162/NECO.1995.7.3.565

BibTeX

@article{dayan1995neco-competition,
  title     = {{Competition and Multiple Cause Models}},
  author    = {Dayan, Peter and Zemel, Richard S.},
  journal   = {Neural Computation},
  year      = {1995},
  pages     = {565-579},
  doi       = {10.1162/NECO.1995.7.3.565},
  volume    = {7},
  url       = {https://mlanthology.org/neco/1995/dayan1995neco-competition/}
}