Ngoc Nguyen

Welcome! I am an Assistant Professor at the Department of Computer Science at University of Dayton (UD), OH. My research is broadly connected to the fields of artificial intelligence (AI) and machine learning, human-computer interaction, cognitive science, and computational social sciences. I use both experimental and computational approaches to understand how people learn to make decisions under uncertainty and how to design AI-powered systems that effectively support human decision-making processes and are helpful for humans to collaborate with when making joint decisions.

Prior to UD, I spent three years as a Postdoctoral Fellow at Carnegie Mellon University’s Dynamic Decision Making Lab (DDMLab) and six months as a Visiting Faculty Lecturer at the Miner School of Computer & Information Sciences, University of Massachusetts Lowell. I hold a Ph.D. in Computer Science from the Faculty of Computer Science at the Free University of Bozen-Bolzano, Italy, where my research focused on supporting group decision-making processes with conversational recommendation techniques.

Research Interests: My research interests include personalized, adaptive systems; computational modeling of human behavior; artificial intelligence; human-AI interaction; and machine learning.

News

06/2025: My NSF CRII proposal was funded by NSF. Thanks NSF!

06/2025: Our book chapter titled 'Credit assignment: challenges and opportunities in developing human-like learning agents' has been published in Bi-directionality in Human-AI Collaborative Systems - Elsevier ISBN: 9780443405532.

05/2025: Full paper titled 'Measuring Implicit Spatial Coordination in Teams: Effects on Collective Intelligence and Performance' has been accepted for publication at the ACM Collective Intelligence Conference 2025 (CI'25).

04/2025: Journal article titled 'Federated learning with randomized alternating direction method of multipliers and application in training neural networks' has been published in the journal Neural Networks.

03/2025: Our proposal NSF STTR Phase I was funded. Thanks NSF!

01/2025: Awarded the UD Research Council Seed Grant (competitive internal research grant)

08/2024: Full paper titled "Predicting and Understanding Human Action Decisions: Insights from Large Language Models and Cognitive Instance-Based Learning" has been accepted at the AAAI HCOMP 2024.

03/2024: Awarded UD/UDRI Summer Research Fellowship

02/2024: Awarded the UD Research Council Seed Grant (competitive internal research grant)

Research Projects

Cognitive and Machine Learning Models for Cooperative Multiagent Systems image

Cognitive and Machine Learning Models for Cooperative Multiagent Systems Developing effective multi-agent systems (MAS) is crucial for applications requiring collaboration with humans. We propose multi-agent Instance-based Learning (MAIBL) models that integrate cognitive mechanisms with multi-agent deep reinforcement learning (MADRL) techniques, demonstrating faster learning and improved coordination in dynamic, stochastic environments. [Code] [Paper]

Interactive Decision Making Games for Behavioral and Human-AI Research image

Interactive Decision Making Games for Behavioral and Human-AI Research In this work, we developed a dynamic interactive game for multi-step goal-directed decision making tasks. It offers customization, experimental scenario creation, and visualization tools for analyzing human decision-making. [Code] [Demo] [Paper] [Interactive Gridworld] [One-step Gridworld]

Credit Assignment for Developing Human-like Learning Agents image

Credit Assignment for Developing Human-like Learning Agents We used a cognitive model based on Instance-Based Learning Theory (IBLT) and Temporal Difference techniques to investigates various credit assignment mechanisms and compares them with human decision-making. [Code] [Demo] [Paper]

Visualizing Human Strategies in Minecraft Search and Rescue Missions image

Visualizing Human Strategies in Minecraft Search and Rescue Missions This project processes data to visualize human participants playing a search and rescue task in Minecraft on a 2D map, aiming to learn about human behavior structures in complex navigation tasks. [Code] [Demo]

A chat-based group recommender system image

A chat-based group recommender system This project developed a chat-based group recommender system for suggesting points of interest (POIs) to groups, implemented as an Android mobile application. It explored how an individual's conflict resolution style affected their receptivity to group recommendations. [Demo]

[Go to top]