A Potts spin MFT network solving multiple causal interactions

Authors:
Lotfi Ben Romdhane
Affiliations:
RIFA, D.M.I, Faculty of Sciences, University of Sherbrooke, Sherbrooke, QC, Canada
Venue:
IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
Year:
1999

Citing 4
Cited 0

The computational complexity of abduction

Artificial Intelligence - Special issue on knowledge representation
Connectionism and explanation

TINLAP '87 Proceedings of the 1987 workshop on Theoretical issues in natural language processing
A unified model for abduction-based reasoning

IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
Optimal broadcast scheduling in packet radio networks using mean field annealing

IEEE Journal on Selected Areas in Communications

Quantified Score

Hi-index	0.00

Visualization

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.