A Potts Spin MFT Network Solving Multiple Causal Interactions
Abstract
In this paper, we propose a Potts spin Mean Field annealed network to address the open, independent and incompatibility classes of causal reasoning (also said abduction, abductive diagnosis). The strong feature of the current work is its characterization of the reasoning task in these classes by an energy/target function. Computation of a scenario (also said explanation) is done by means of Mean Field equations. The application of the model to small and large-scale causal problems reveals its efficacy and robustness in handling varied and multiple causal interactions.
Cite
Text
Romdhane. "A Potts Spin MFT Network Solving Multiple Causal Interactions." International Joint Conference on Artificial Intelligence, 1999.Markdown
[Romdhane. "A Potts Spin MFT Network Solving Multiple Causal Interactions." International Joint Conference on Artificial Intelligence, 1999.](https://mlanthology.org/ijcai/1999/romdhane1999ijcai-potts/)BibTeX
@inproceedings{romdhane1999ijcai-potts,
title = {{A Potts Spin MFT Network Solving Multiple Causal Interactions}},
author = {Romdhane, Lotfi Ben},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {1999},
pages = {814-819},
url = {https://mlanthology.org/ijcai/1999/romdhane1999ijcai-potts/}
}