Research Direction
We aim to build reliable AI systems that quantify uncertainty in new data and use it to quickly adapt to changing environments. We are also interested in adaptive learning for multimodal data and in developing AI methods that incorporate physical laws to model real-world physical systems. Our research interests and future directions include:
Continual and Adaptive Learning
Learning new dataset quickly and efficiently by leveraging previously learned knowledge.
Uncertainty-Aware Learning
Quantifying uncertainty to represent what AI models know and don't know.
Multimodal Foundation Models
Developing foundation models for learning and inference across multi-modality data.
Scientific AI for Physical Systems
Developing AI methods that incorporate physical laws to model real-world physical systems.
News
Awarded a two-year research grant (Core Basic Research B) from the National Research Foundation (NRF).
Gave special lectures at the Department of Mathematics, Ewha Womans University.
Attended the 4th Bayes-Duality Workshop at RIKEN AIP.
Started Reliable Adaptive Intelligence Lab at Jeonbuk National University.
Join Us
- RAI Lab is looking for motivated undergraduate interns and graduate students from computer science, statistics, mathematics, and related fields who are interested in reliable, adaptive, and uncertainty-aware AI.
- If undergraduate students (Years 2–3) are interested in a research internship, please read the research intern curriculum and apply using this form.
- Master’s applicants may email yohan.jung [at] jbnu [dot] ac [dot] kr with a CV; positions for international students are currently unavailable.