From Moral Reasoning to Visual Reasoning: Using AI to Model and Understand Intelligent Decision Making
March 13, 2020 | 12pm
From Moral Reasoning to Visual Reasoning: Using AI to Model and Understand Intelligent Decision Making
My research aims to investigate how humans or other intelligent agents make meaningful decisions, and how to enhance human-technology partnerships through the development of AI systems that promote inclusion and assist people. In particular, my research methods focus on combining cognitively inspired computational approaches with reinforcement learning (RL) techniques. In this talk, I introduce two cognitively inspired computational architectures that apply RL to different domains: moral reasoning and visual reasoning. First, I describe a multi-agent architecture that uses artificial sensations, emotions, and feelings to model moral reasoning and that illustrates the emergence of cooperation in multi-agent systems. Second, I discuss a computational architecture designed to investigate how humans perform open-ended visual data exploration.
About Fernanda
Fernanda M. Eliott is a Data Science Institute (DSI) postdoctoral fellow affiliated with the Department of Electrical Engineering and Computer Science at Vanderbilt University. She is a member of the Artificial Intelligence and Visual Analogical Systems (AIVAS) lab, the Vanderbilt Initiative in Data-intensive Astrophysics (VIDA), and the Frist Center for Autism and Innovation. Her research in artificial intelligence and cognitive 
science studies the roles of perception, memory, emotions, and feelings in the processes of human decision making. Dr. Eliott earned her M.S. and Ph.D. degrees in Sciences from the Aeronautics Institute of Technology (ITA, Brazil), working with the Artificial Intelligence and Robotics Group (AIRGroup), and her Bachelor’s and Licentiate’s degrees in Philosophy from the University of São Paulo (USP).