Hi, I'm Jaesik(재식)! I am working to explore a variety of new opportunities to extend the generality of Artificial Intelligence (AI) besides developing AI services.
I'm a senior machine learning developer at SAP, and a Ph.D. student under the supervision of Prof. Sungjin Ahn at MLML. Before that, I worked in Samsung Electronics and graduated from KAIST.
We've studied
learning-to-learn (a.k.a meta-learning or few-shot learning) [c2, p1, c3, c4],
unsupervised learning of compositional representation such as object-centric representation [c5, c6],
and agent learning [w1, c6, c7, c8].
I am interested in the diverse ways to extend the generality of AI such as compositional representation or the agent learning with more extendable architectures.
Publications:
(c: peer-reviewed conference, w: peer-reviewed workshop, p: preprint, * = equal contributions)
[c8] Dr. Strategy: Model-Based Generalist Agents with Strategic Dreaming (arXiv)
H. Hamed*, S. Kim*, D. Kim, J. Yoon, S. Ahn
International Conference on Machine Learning (ICML) 2024
[c7] Spatially-Aware Transformers for Embodied Agents (arXiv, site)
J. Cho* and J. Yoon*, S. Ahn
International Conference on Learning Representation (ICLR) 2024
Spotlight Presentation (top 5%)
[c6] An Investigation into Pre-Training Object-Centric Representations for Reinforcement Learning (arXiv, site, code, data, logs)
J. Yoon, Yi-Fu Wu, H. Bae, S. Ahn
International Conference on Machine Learning (ICML) 2023
[w1] TransDreamer: Reinforcement Learning with Transformer World Models (arXiv)
C. Chen*, Yi-Fu Wu*, J. Yoon*, S. Ahn
Deep RL NeurIPS 2021
[c5] Generative Video Transformer: Can Objects be the Words? (arXiv)
Yi-Fu Wu, J. Yoon, S. Ahn
International Conference on Machine Learning (ICML) 2021
[c4] Robustifying Sequential Neural Processes (arXiv, code)
J. Yoon, G. Singh, S. Ahn
International Conference on Machine Learning (ICML) 2020
[c3] Sequential Neural Processes (arXiv, site, code)
G. Singh*, J. Yoon*, Y. Son, S. Ahn
Conference on Neural Information Processing Systems (NeurIPS) 2019
Spotlight Presentation (top 2.4% = 163/6743)
[p1] One-Shot Learning for Text-to-SQL Generation (arXiv)
D. Lee, J. Yoon, J. Song, S. Lee, S. Yoon
Preprint
[c2] Bayesian Model-Agnostic Meta-Learning (arXiv, code, talk)
J. Yoon*, T. Kim*, O. Dia, S. Kim, Y. Bengio, S. Ahn
Conference on Neural Information Processing Systems (NeurIPS) 2018
Spotlight Presentation (top 3.5% = 168/4856)
[c1] A hierarchical-structured dictionary learning for image classification (arXiv)
J. Yoon, J. Choi, CD Yoo
International Conference on Image Processing (ICIP) 2014
Oral Presentation
Services
Have serviced as a reviewer from 2021 for NeurIPS, ICML, and ICLR.