Recognizing Concepts and Recognizing Musical Themes. A Quantum Semantic Analysis

TL;DR

This paper explores recognizing abstract concepts and musical themes using a quantum pattern recognition framework, comparing human and AI behaviors, and proposing quantum data sets to model uncertainties and Gestalt-like patterns.

Contribution

It introduces a quantum approach to pattern recognition that models Gestalt formation and classification of musical themes and concepts in both humans and machines.

Findings

Quantum data sets represent ambiguous patterns.

Quantum similarity relations aid classification.

Both humans and AI can use quantum-inspired methods.

Abstract

How are abstract concepts and musical themes recognized on the basis of some previous experience? It is interesting to compare the different behaviors of human and of artificial intelligences with respect to this problem. Generally, a human mind that abstracts a concept (say, table) from a given set of known examples creates a table-Gestalt: a kind of vague and out of focus image that does not fully correspond to a particular table with well determined features. A similar situation arises in the case of musical themes. Can the construction of a gestaltic pattern, which is so natural for human minds, be taught to an intelligent machine? This problem can be successfully discussed in the framework of a quantum approach to pattern recognition and to machine learning. The basic idea is replacing classical data sets with quantum data sets, where either objects or musical themes can be formally represented as pieces of quantum information, involving the uncertainties and the ambiguities that characterize the quantum world. In this framework, the intuitive concept of Gestalt can be simulated by the mathematical concept of positive centroid of a given quantum data set. Accordingly, the crucial problem "how can we classify a new object or a new musical theme (we have listened to) on the basis of a previous experience?" can be dealt with in terms of some special quantum similarity-relations. Although recognition procedures are different for human and for artificial intelligences, there is a common method of "facing the problems" that seems to work in both cases.