Polarization in agent based models: learning in neural networks and emerging collective properties

• Grant number: 19/27825-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2020
• End date: October 31, 2021
• Field of knowledge: Physical Sciences and Mathematics - Physics - General Physics
• Principal Investigator: Nestor Felipe Caticha Alfonso
• Grantee: José Arthur de Toledo Queiroz
• Host Institution: Instituto de Física (IF). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

The student will participate in the research program that deals with neural networks learning and the construction and characterization of agents that process information using neural networks. The theoretical bases are Probability Theory, Statistical Mechanics, Information Theory and Machine Learning. The methods include analytical and computational techniques. Societies of agents will be built to model Polarization processes with respect to opinions. After a period of study of the theoretical bases, the student should build models and make machine comparisons by learning a task using different training methods and confronting theoretical predictions for models of societies with large opinion databases in various areas. such as voting patterns of legislative systems in different countries, morality and digital social networks.