Emergence of symbolic behavior from brain like memory with dynamic attention

Omori, Takashi; Mochizuki, Akiko; Mizutani, Kentaro; Nishizaki, Makoto

doi:10.1016/s0893-6080(99)00054-4

Cited by 23 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Incidentally, a neural network model named PATON [11,12] has been proposed. This model can perform context-dependent association and realize the same state transitions as the automaton, and is similar to the proposed model in that the output gain of some of the elements is set to 0, depending on the context.…”

Section: Discussionmentioning

confidence: 99%

A model of context-dependent association using selective desensitization of nonmonotonic neural elements

2005

View full text Add to dashboard Cite

SUMMARYIn neural networks based on a distributed information representation, when different patterns are to be recalled from the same input by association depending on the context, the usual method is to concatenate the pattern representing the context with the input pattern. However, there is a serious problem in this approach, and strong constraints are imposed on the number of inputs and the number of context patterns. This paper tries to apply a different method of contextual modification to nonmonotonic neural networks in order to construct a context-dependent associative model that can solve the longstanding problem. In the proposed model, the number of associations that can be learned increases almost in proportion to the number of elements, regardless of the numbers of the input and context patterns. In addition, the state transitions among attractors can be controlled flexibly by switching the context, which enables the model to simulate the behavior of any finite automaton without inducing an explosive increase in the number of elements or the training time. The model also has a high generalization power ability based on fully distributed representations and has potential for overcoming the limitations of conventional symbol processing.

show abstract

Section: Discussionmentioning

confidence: 99%

A model of context-dependent association using selective desensitization of nonmonotonic neural elements

2005

View full text Add to dashboard Cite

show abstract

“…In PATON proposed by Omori et al [112], [113], the behavior of recognition and association in an associative memory system is controlled by a contextdependent switch. In motion control, Wolpert and Kawato [114], [115] prepared multiple neural-network modules.…”

Section: Development Context Dependence Volition and Developmentalmentioning

confidence: 99%

Complex-Valued Neural Networks

Hirose¹

2006

Studies in Computational Intelligence

150

View full text Add to dashboard Cite

“…Please refer to our previous presentation [7] for more detailed discussions on this topic. Compared to other researches in similar motivations [6], [11], [13], [14], our method has advantages in that it can deal with a large number of high dimensional patterns and it can self-organize conceptual-specific hierarchy based on the structure of stored patterns.…”

Section: A Memory Of Motions For Humanoid Robotsmentioning

confidence: 99%

Hierarchical Concept Formation in Associative Memory Models and its Application to Memory of Motions for Humanoid Robots

Kadone

Nakamura

2006

2006 6th IEEE-RAS International Conference on Humanoid Robots

View full text Add to dashboard Cite

In associative neural networks with nonmonotonic activation functions which store hierarchically correlated patterns, bifurcations of attractors take place depending on the parameter of nonmonotonicity. With hierarchically correlated storage patterns, attractors are the stored patterns when the nonmonotonicity is large, and new emergent patterns at around the centers of clusters when the nonmonotonicity is small. The phenomenon itself was shown by the authors[7] by simulations and applied to memory systems for humanoid robots, which store feature vectors of motion patterns and maintain specific and emergent conceptual representations of motions of humanoid robots. In this paper, we theoretically describe the hierarchical bifurcations of attractors in nonmonotonic associative memory models and discuss the correspondences between the theory and simulation results.

show abstract

Emergence of symbolic behavior from brain like memory with dynamic attention

Cited by 23 publications

References 11 publications

A model of context-dependent association using selective desensitization of nonmonotonic neural elements

A model of context-dependent association using selective desensitization of nonmonotonic neural elements

Complex-Valued Neural Networks

Hierarchical Concept Formation in Associative Memory Models and its Application to Memory of Motions for Humanoid Robots

Contact Info

Product

Resources

About