Stability analysis of attractor neural network model of inferior temporal cortex - Relationship between attractor stability and learning order

摘要

Miyashita found that the long-term memory of visual stimuli is stored in the monkey's inferior temporal cortex and that the temporal correlation in terms of the learning order of visual stimuli is converted into spatial correlation in terms of the.ring rate patterns of the neuron group. To explain Miyashita's.ndings, Griniasty et al. [Neural Comput. 5(1993) 1] and Amit et al. [J. Neurosci. 14(1994) 6435] proposed the attractor neural network model, and the Amit model has been examined only for the stable state acquired by storing memory patterns in a.xed sequence. In the real world, however, the learning order has statistical continuity but it also has randomness, and the stability of the state changes depending on the statistical properties of learning order when memory patterns are stored randomly. In addition, it is preferable for the stable state to become an appropriate attractor that re.ects the relationship between memory patterns by the statistical properties of the learning order. In this study, we examined the dependence of the stable state on the statistical properties of the learning order without modifying the Amit model. The stable state was found to change from the correlated attractor to the Hop.eld or M _p attractor, which is the mixed state with all memory patterns when the rate of random learning increases. Furthermore, we found that if the statistical properties of the learning order change, the stable state can change to an appropriate attractor re.ecting the relationship between memory patterns.