About

I am a Ph.D. student of Mathematical Sciences at Seoul National University, advised by Professor Ernest K. Ryu. My current interest is in generative models, especially diffusion probabilistic models and large language models.

---

Publications

LoRA can Replace Time and Class Embeddings in Diffusion Probabilistic Models Joo Young Choi, Jaesung Park, Inkyu Park, Jaewoong Cho, Albert No, Ernest K. Ryu. NeurIPS 2023 Workshop on Diffusion Models

Diffusion Probabilistic Models Generalize when They Fail to Memorize TaeHo Yoon, Joo Young Choi, Sehyun Kwon, Ernest K. Ryu. ICML 2023 Workshop SPIGM

Rotation and Translation Invariant Representation Learning with Implicit Neural Representations Sehyun Kwon, Joo Young Choi, Ernest K. Ryu. ICML 2023

Neural Tangent Kernel Analysis of Deep Narrow Neural Networks Jongmin Lee, Joo Young Choi, Ernest K. Ryu, Albert No. ICML 2022

---

Education

Seoul National University (September 2018 - )
Ph.D in Mathematical Sciences

Korea University (March 2012 - August 2018)
B.S in Business Administration
B.S in Mathematics

---

Experience

Teaching Assistant (Seoul National University)

• Topics in Machine Intelligence: Generative AI and Foundation Models, M3309.001800, Spring 2024.
• Mathematical Foundations of Deep Neural Networks, M1407.001200, Fall 2022. Outstanding TA Award
• Topics in Applied Mathematics: Infinitely Large Neural Networks, 3341.751, Spring 2022.
• Mathematical Foundations of Deep Neural Networks, M1407.001200, Fall 2021.

---

Notes

Study of papers that are of personal interest

• T-MARS: Improving Visual Representations by Circumventing Text Feature Learning
  Slides / Paper
• What Do We Learn from Inverting CLIP Models?
  Slides / Paper
• GLIGEN: Open-Set Grounded Text-to-Image Generation
  Slides / Paper
• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
  Slides / Paper
• Tree of Thoughts: Deliberate Problem Solving with Large Language Models
  Slides / Paper
• Visual Instruction Tuning
  Slides / Paper
• Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
  Slides / Paper
• LoRA: Low-Rank Adaptation of Large Language Models
  Slides / Paper
• Visual Prompting via Image Inpainting
  Slides / Paper
• Trainig Data Attribution for Diffusion Models
  Slides / Paper
• Optimizing DDPM Sampling with Shortcut Fine-Tuning
  Slides / Paper
• Unsupervised Representation Learning from Pre-trained Diffusion Probabilistic Models
  Slides / Paper
• SELF-REFINE: Iterative Refinement with Self-Feedback
  Slides / Paper
• Generative Agents: Interactive Simulacra of Human Behavior
  Slides / Paper
• Consistency Models
  Slides / Paper
• Image as Set of Points
  Slides / Paper
• Traditional Classification Neural Networks are Good Generators: They are Competitive with DDPMs and GANs
  Slides / Paper
• Git Re-Basin: Merging Models modulo Permutation Symmetries
  Slides / Paper