Development of a Mathematical Apparatus with an Imagery Representation of Information for Neuromorphic Systems
Nikolai A. Simonov · Russian Microelectronics · 2023
Abstract In this work, we consider the possibility of using a mathematical model of spots for the representation of secondary mental imagery in psychology and imagery representation of information in AI, as well as for solving inverse problems, including application of a learning method. A basic mathematical apparatus has been developed for the presentation and processing of qualitative information about elementary spatial relations between spots. The proposed apparatus can be used to develop algorithms for neuromorphic systems based on new promising components of memristors and FeFETs, which have nonvolatile memory, extremely low switching losses and faster read/write operations. It provides information representation and processing in the imagery form.