Research
Papers
Fact AI Lab is built on applied research. The work below - published at peer-reviewed venues including IROS, AAAI, ICRA, and Elsevier journals - covers the core technical challenges behind reliable AI: uncertainty quantification, multi-modal evaluation, verifiable reasoning, and system behavior under distribution shift.
2026
Beyond Static Datasets: Robust Offline Policy Optimization via Vetted Synthetic Transitions
P. Agand, M. Chen·arXiv preprint
2025
Recommender System for Data Science Learning and Research
P. Agand, et al.·International Journal of Artificial Intelligence in Teaching and Learning (IJAITL)
2024
Sequential Modeling of Complex Marine Navigation
P. Agand, et al.·AAAI
DMODE: Differential Monocular Object Distance Estimation Without Class-Specific Information
P. Agand, M. Chang, M. Chen·13th International Workshop on Robot Motion and Control (RoMoCo)
DMFuser: Distilled Multi-Task Learning for Transformer-Based Multi-modal Fusion
P. Agand, et al.·IROS
DaTu: AI-Assisted Exploratory Data Analysis for Data Science Education
P. Agand, et al.·Software Impacts, Elsevier
2023
LeTFuser for Autonomous Driving with Multi-Task Learning
P. Agand, et al.·CVPR Workshop on Autonomous Vehicles
Fuel Consumption Prediction for a Ferry using ML and In-service Data
P. Agand, et al.·Ocean Engineering
DRL Traffic Signal Controls with Optimized CO2 Emissions
P. Agand, et al.·IROS
Online Probabilistic Model Identification Using Adaptive Recursive MCMC
P. Agand, M. Chen, H.D. Taghirad·International Joint Conference on Neural Networks (IJCNN)
2022
2021
EcoLight: Reward Shaping in DRL for Environment Friendly Traffic Signal Control
P. Agand, et al.·NeurIPS Workshop on Tackling Climate Change with Machine Learning