Noisy-or nodes for conditioning models

Authors:
Jean Marc Salotti
Affiliations:
Groupe Cognitique et Ingénierie Humaine, Ecole Nationale Supérieure de Cognitique, Bordeaux Cedex, France
Venue:
SAB'10 Proceedings of the 11th international conference on Simulation of adaptive behavior: from animals to animats
Year:
2010

Citing 8
Cited 0

Probabilistic reasoning in intelligent systems: networks of plausible inference

Probabilistic reasoning in intelligent systems: networks of plausible inference
The computational complexity of probabilistic inference using Bayesian belief networks (research note)

Artificial Intelligence
A Bayesian Method for the Induction of Probabilistic Networks from Data

Machine Learning
Computational models of classical conditioning: a comparative study

Proceedings of the fifth international conference on simulation of adaptive behavior on From animals to animats 5
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Qualitative Analysis of Behavior, Vol. 11: Neural Network Models of Conditioning and Action

Qualitative Analysis of Behavior, Vol. 11: Neural Network Models of Conditioning and Action
Representation and timing in theories of the dopamine system

Neural Computation
Multiple model-based reinforcement learning explains dopamine neuronal activity

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In order to adapt the behavior of robots to varying environments, conditioning models provide interesting ideas. A prediction system is an important part of such models. The problem is to update it according to the sequence of stimuli perceived by the robot. Bayesian networks can be used to implement the prediction system. However, update rules are very complex and we need an incremental and fast learning process. We propose the use of noisy or nodes with appropriate learning rules. Numerous features of conditioning have been tested and promising results have been obtained.