In our real life, it is well known that our cognitive process is always influenced by our environment. It is called as “context dependency” of the cognition. In this paper, we propose a memory model “PATON” that is based on a macroscopic structure of a cortico-hippocampal memory system; it has three components of a symbolic layer, a pattern layer, and an attentional system. The attentional system sends signals to control a change of the model's structure dynamically. The change induces a modulation of metric between memorized items. Computer simulation shows an association process dependent upon a context based on the modulation.
Pattern recognition invariant against deformation or translation can be performed with dynamic link architecture, which has been proposed by von der Malsburg. Dynamic link is applied to many engineering examples efficiently, but has not been analyzed mathematically. We propose a mathematically tractable model of dynamic link architecture. Our model matches common parts between a template image and a data image. To analyze model mathematically, we derive the phase dynamics from the model equation. We also carry out a computer simulation to verify the mathematical theory.
Our visual sensor consists of many retinal neurons, each of which covers only a local receptive field. In order to recognize the object location as an analog value, we have to integrate visual signals from many retinal neurons. In our real world, objects usually move smoothly according to the equations of motion. If an output of a neural network which receives inputs from retinal neurons, changes smoothly along time, the output represents the object location as an analog value. Then the learning method has been proposed, which makes the output of the neural network become smooth, in other words, makes the second time derivative of the output become 0.