@Min Gu Kwak

📍Table of Contents📍

I am a Machine Learning Research Scientist at the University of Pittsburgh and an Affiliate Researcher at Georgia Institute of Technology. My research focuses on developing artificial intelligence models for healthcare applications across vision and language domains.

At the University of Pittsburgh, I work on Large Language Models (LLMs) for clinical data standardization and electronic health records, leading the development of ReDWINE clinical data warehouse with LLM-based automated mapping systems.

My work at Georgia Tech has centered on generative models and medical imaging analysis, including applications in Alzheimer's disease diagnosis, brain tumor characterization, and dental lesion detection. Through bridging theoretical AI advancements and practical clinical needs, I aim to develop systems that enhance healthcare delivery and improve patient outcomes.

🔗 [[LinkedIn]](https://https://www.linkedin.com/in/min9kwak/) [Google Scholar]

🏢 Office: #6025 Forbes Tower 📨 E-mail: [email protected]

🖊️ Publications

📚 Google Scholar

📚 Published Papers

• Safe Semi-Supervised Contrastive Learning Using In-Distribution Data as Positive Examples Kwak, M. G., Kahng, H., & Kim, S. B. IEEE Access (2025) [link] 🤝 Semi-Supervised 🧲 Contrastive Learning 🛡️ Out-of-Distribution Robustness

• A Cross-Modal Mutual Knowledge Distillation Framework for Alzheimer’s Disease Diagnosis: Addressing Incomplete Modalities Kwak, M. G., Mao, L., Zheng, Z., Su, Y., Lure, F., & Li, J. IEEE Transactions on Automation Science and Engineering (2025) [link] 🧠 Alzheimer’s Diagnosis 🔁 Knowledge Distillation 🧩 Incomplete Multi-Modality

• Mixing corrupted preferences for robust and feedback-efficient preference-based reinforcement learning Heo, J., Lee, Y. J., Kim, J., Kwak, M. G., Park, Y. J., & Kim, S. B. Knowledge-Based Systems (2025) [link] 🎛️ Preference-Based RL 🧩 Corrupted Feedback Robustness ⚡ Feedback Efficiency

• Oral-Anatomical Knowledge-Informed Semi-Supervised Learning for 3D Dental CBCT Segmentation and Lesion Detection Lee, Y., Kwak, M. G., Chen, R. Q., Yan, H., Mupparapu, M., Lure, F., ... & Li, J. ****IEEE Transactions on Automation Science and Engineering (2025) [link] 🤝 Semi-Supervised 🧩 Knowledge Integration ⚡ Image Segmentation

• DynaSTI: Dynamics modeling with sequential temporal information for reinforcement learning in Atari Kim, J., Lee, Y. J., Kwak, M., Park, Y. J., & Kim, S. B. Knowledge-Based Systems (2024) [link] 🎮 Reinforcement Learning 🔄 Hierarchical Dynamics Modeling ⚡ Sample Efficiency

• Masked and inverse dynamics modeling for data-efficient reinforcement learning Lee, Y. J., Kim, J., Park, Y. J., Kwak, M., & Kim, S. B. IEEE Transactions on Neural Networks and Learning Systems (2024) [link] 🎮 Reinforcement Learning 🎭 Masked & Inverse Dynamics Modeling ⚡ Data Efficiency

• Self-supervised contrastive learning to predict the progression of Alzheimer’s disease with 3D amyloid-PET Kwak, M. G., Su, Y., Chen, K., Weidman, D., Wu, T., Lure, F., & Li, J. Bioengineering (2023) **[link] 🧠 Alzheimer’s Disease Progression 🧲 Self-Supervised Contrastive Learning 🧬 3D Image