Modeling agent decision and behavior in the light of data science and artificial intelligence

• Agent-based modeling (ABM) has been widely used in numerous disciplines and practice domains, subject to many eulogies and criticisms. This article presents key advances and challenges in agent-based modeling over the last two decades and shows that understanding agents' behaviors is a major priority for various research fields. We demonstrate that artificial intelligence and data science will likely generate revolutionary impacts for science and technology towards understanding agent decisions and behaviors in complex systems. We propose an innovative approach that leverages reinforcement learning and convolutional neural networks to equip agents with the intelligence of self-learning their behavior rules directly from data. We call for further developments of ABM, especially modeling agent behaviors, in the light of data science and artificial intelligence.

Ämnesord

TEKNIK OCH TEKNOLOGIER  -- Naturresursteknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Environmental Engineering (hsv//eng)

NATURVETENSKAP  -- Geovetenskap och miljövetenskap (hsv//swe)

NATURAL SCIENCES  -- Earth and Related Environmental Sciences (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datavetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Sciences (hsv//eng)

Nyckelord

Agent-based modeling

Modeling agent decisions and actions

Artificial intelligence

Machine learning

Data science

Publikations- och innehållstyp

ref (ämneskategori)

art (ämneskategori)