Younggyo Seo

I am a Ph.D Student at KAIST, advised by Jinwoo Shin, and also a Visiting Ph.D Student at UC Berkeley, advised by Pieter Abbeel and Kimin Lee. I also have collaborated with Honglak Lee at University of Michigan.

Previously, I've interned at Microsoft Research Asia as a research intern in Deep and Reinforcement Learning Group. Before that, I graduated with B.A. in Economics from Seoul National University.

Feel free to send me an e-mail if you want to chat or collaborate with me!

Email  /  CV  /  Google Scholar  /  Twitter

profile photo

July '22  

I'll attend my first in-person conference - ICML 2022, Baltimore. Happy to chat if you're interested in my works, or collaborating with me!

June '22  

New website! Still working on making figures for publications.

My research interest lies in visual reinforcement learning (RL), where an agent operates on visual observations. Specifically, I'm interested in developing representation learning and exploration techniques to improve the sample-efficiency of visual RL algorithms. I am also broadly interested in areas related to RL, including video prediction, offline RL, and generalization in RL. Representative papers are highlighted.
Masked World Models for Visual Control
Younggyo Seo, Danijar Hafner, Hao Liu, Fangchen Liu, Stephen James, Kimin Lee, Pieter Abbeel
Preprint, 2022.
pdf / website / code

We introduce MWM that learns a latent dynamics model on top of an autoencoder trained with convolutional feature masking and reward prediction.

Reinforcement Learning with Action-Free Pre-Training from Videos
Younggyo Seo, Kimin Lee, Stephen James, Pieter Abbeel
International Conference on Machine Learning (ICML), 2022.
pdf / website / code

We introduce APV that can leverage diverse videos from different domains for pre-training to improve sample-efficiency.

HARP: Autoregressive Latent Video Prediction with High-Fidelity Image Generator
Younggyo Seo, Kimin Lee, Fangchen Liu, Stephen James, Pieter Abbeel
International Conference on Image Processing (ICIP), 2022.
pdf / code

We introduce a video prediction model that can generate 256x256 frames by training an autorgressive transformer on top of VQ-GAN.

Object-Aware Regularization for Addressing Causal Confusion in Imitation Learning
Jongjin Park*, Younggyo Seo*, Chang Liu, Li Zhao, Tao Qin, Jinwoo Shin, Tie-Yan Liu
Neural Information Processing Systems (NeurIPS), 2021.
pdf / article / code

We introduce OREO, a regularization technique for behavior cloning from pixels.

Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learning
Junsu Kim, Younggyo Seo, Jinwoo Shin
Neural Information Processing Systems (NeurIPS), 2021.
pdf / code

We introduce HIGL, a goal-conditioned hierarchical RL method that samples landmarks and utilizes them for guiding the training of a high-level policy.

Offline-to-Online Reinforcement Learning via Balanced Replay and Pessimistic Q-Ensemble
Seunghyun Lee*, Younggyo Seo*, Kimin Lee, Pieter Abbeel, Jinwoo Shin
Conference on Robot Learning (CoRL), 2021.
pdf / code

We introduce an offline-to-online RL algorithm to address the distribution shift that arises during the transition between offline RL and online RL.

State Entropy Maximization with Random Encoders for Efficient Exploration
Younggyo Seo*, Lili Chen*, Jinwoo Shin, Honglak Lee, Pieter Abbeel, Kimin Lee
International Conference on Machine Learning (ICML), 2021.
pdf / website / code

We introduce RE3, an exploration technique for visual RL that utilizes a k-NN state entropy estimate in the representation space of a randomly initialized & fixed CNN encoder.

Trajectory-wise Multiple Choice Learning for Dynamics Generalization in Reinforcement Learning
Younggyo Seo*, Kimin Lee*, Ignasi Clavera, Thanard Kurutach, Jinwoo Shin, Pieter Abbeel
Neural Information Processing Systems (NeurIPS), 2020.
pdf / website / code

We introduce T-MCL that learns multi-headed dynamics model whose each prediction head is specialized in certain environments with similar dynamics, i.e., clustering environments.

Context-aware Dynamics Model for Generalization in Model-Based Reinforcement Learning
Kimin Lee*, Younggyo Seo*, Seunghyun Lee, Honglak Lee, Jinwoo Shin
International Conference on Machine Learning (ICML), 2020.
pdf / website / code

We introduce CaDM that learns a context encoder to extract contextual information from recent history with self-supervised losses of future and backward prediction.

Learning What to Defer for Maximum Independent Sets
Sungsoo Ahn, Younggyo Seo, Jinwoo Shin
International Conference on Machine Learning (ICML), 2020.
pdf / code

We introduce LwD, a deep RL framework for the maximum independent set problem.


Visiting Student | UC Berkeley
June 2021 - Present

Working with Pieter Abbeel at the robot learning lab on various research projects.

Research Intern | Microsoft Research Asia
Dec 2020 - May 2021

At Deep and Reinforcement Learning Group, I worked with Chang Liu, Li Zhao, and Tao Qin on a research project that proposed a regularization technique for imitation learning.


Ph.D in Artificial Intelligence | KAIST
Sep 2019 - Aug 2023 (Expected)

B.A in Economics | Seoul National University
Mar 2012 - Feb 2019

  • Summa Cum Laude
  • Leave of absence for mandatory military service: Feb 2014 - Feb 2016

  • Conference reviewer: ICLR, ICML, NeurIPS, CoRL
  • Journal reviewer: TMLR

  • Top reviewer award (top 10%), ICML 2021, 2022
  • Best paper award, Korean Artificial Intelligence Association, 2021
  • Summa Cum Laude, Economics department, Seoul National University, 2019
  • National Humanities Scholarship, Korea Student Aid Foundation, 2012-2018
  • 1st Rank @ College Scholastic Ability Test with perfect score (500/500), 2011


Website templates from here and here.