Young-Jin Park

I’m PhD Student at MIT LIDS, advised by Prof. Navid Azizan. My current research interest lies in quantifying reliability of foundational models (e.g., LLMs).

I was an AI research engineer at NAVER AI Lab & CLOVA. At NAVER, I developed a user behavior models and recommender systems using large language models (LLMs) for tens of millions of users. I also conducted research on large-scale representation learning for temporal data (e.g., demand forecasting systems).

I received my M.S. and B.S. degree in Aerospace Engineering and Mathematical Sciences (minor) from KAIST in 2019 and 2017, respectively, under the supervision of Prof. Han-Lim Choi. During my master’s degree, I researched 1) hierarchical RL and 2) deep latent variable models , aiming to bridge the gap between high-level reasoning and low-level control policies.

Selected Publications

  1. Representation Reliability and Its Impact on Downstream Tasks
    Young-Jin Park, Hao Wang, Shervin Ardeshir, and Navid Azizan
    arXiv preprint 2023
  2. A Large-Scale Ensemble Learning Framework for Demand Forecasting
    Young-Jin Park, Donghyun Kim, Frédéric Odermatt, Juho Lee, and Kyung-Min Kim
    In IEEE ICDM 2022 (Full Paper, Acceptance Rate: 9.77%)
  3. Distilling a hierarchical policy for planning and control via representation and reinforcement learning
    Jung-Su Ha*, Young-Jin Park*, Hyeok-Joo Chae, Soon-Seo Park, and Han-Lim Choi
    In IEEE ICRA 2021
  4. A Worrying Analysis of Probabilistic Time-series Models for Sales Forecasting
    Seungjae Jung*, Kyung-Min Kim*, Hanock Kwak*, and Young-Jin Park*
    In NeurIPS Workshop 2020
    🏆 Best Poster Awards
  5. Adaptive Path-Integral Autoencoders: Representation Learning and Planning for Dynamical Systems
    Jung-Su Ha, Young-Jin Park, Hyeok-Joo Chae, Soon-Seo Park, and Han-Lim Choi
    In NeurIPS 2018
  6. Deep gaussian process-based bayesian inference for contaminant source localization
    Young-Jin Park, Piyush M Tagade, and Han-Lim Choi
    IEEE Access 2018 [IF: 4.098]
    (presented in UAI Workshop 2018 as well)