Assistant Research Professor

Department: Institute for Applied AI Innovation

Dr. Kyle Lucke is an Assistant Research Professor in the Institute for Applied AI Innovation at The University of Texas at El Paso. His research focuses on uncertainty quantification for deep learning models used in high-stakes applications. He received his Ph.D. in Computer Science from the University of Idaho in 2026, where he conducted research in the Machine Intelligence and Data Analytics (MIDA) Laboratory. He earned his B.S. in Computer Science, with a minor in Mathematics, from the University of Montana in 2019, graduating summa cum laude and receiving the President’s Outstanding Senior award, and his M.S. in Computer Science from the University of Montana in 2021. His master’s thesis evaluated protein inference methods in shotgun proteomics and contributed a protein standard dataset, an ensemble evaluation engine, and several new inference methods.

Dr. Lucke’s doctoral research addressed the overconfidence and poor calibration of deep learning classifiers. His article “Soft-Label Supervised Meta-Model with Adversarial Samples for Uncertainty Quantification” (Computers, 2025) proposed a meta-model framework that detects misclassifications by combining a soft-label formulation of prediction uncertainty with training on adversarial samples to offset dataset imbalance. His work in biomedical image analysis includes “A2DMN: Anatomy-Aware Dilated Multiscale Network for Breast Ultrasound Semantic Segmentation” (IEEE International Symposium on Biomedical Imaging, 2024), and he co-authored a study of query-efficient adversarial attacks on breast ultrasound classifiers. During his master’s studies, he also co-authored work on efficient selection algorithms using layer-ordered heaps, with applications in proteomics.

Following his doctoral studies, Dr. Lucke held an internship at Idaho National Laboratory (INL). He has served as a graduate teaching assistant at the University of Montana and the University of Idaho, including for courses in introductory machine learning and software design.

Keywords of Expertise: Uncertainty Quantification, Robust AI, Adversarial Machine Learning, Physics-Informed Neural Networks, Biomedical Image Segmentation, Generative Models for Cybersecurity.