ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-09242020-143706


Tipo di tesi
Tesi di laurea magistrale
Autore
ARIOLI, DANIELE
URN
etd-09242020-143706
Titolo
CognAC: a cognitive architecture based on Information Dynamics of Thinking
Dipartimento
INFORMATICA
Corso di studi
INFORMATICA
Relatori
relatore Prof. Bacciu, Davide
relatore Prof. Gervasi, Vincenzo
Parole chiave
  • artificial intelligence
  • RNN
  • cognitive architecture
  • IDyOT
  • music generation
Data inizio appello
09/10/2020
Consultabilità
Tesi non consultabile
Riassunto
The Information Dynamics of Thinking, IDyOT for short, is an implementation of the Global Workspace theory. It can be considered the successor of the Information Dynamics of Music (IDyOM) by Pearce, which had good results in the pitch prediction task, showing that it is capable of predicting human behavior. IDyOT model is presented as a minimal cognitive architecture that, with simple but powerful mathematical principles, can go in accounting for human cognition and creativity and being useful in the explanation of it. While different studies exist, they show how the IDyOT model is in a development phase: in recent years the theory has been refined with new ideas and expansions, but there are only two known partial implementations. In this thesis, it is presented an implementation of a cognitive architecture based on the IDyOT model designed to generate music, defining, and test the components of the model.
File