About me

My name is Ju-Seung Byun. I am a forth-year computer science & engineering Ph.D. student at The Ohio State University advised by Andrew Perrault. My research interests are in deep reinforcement learning, especially focusing on hierarchical structure and distributional reinforcement learning.

Before I came to OSU, I received my Master’s degree from University of Southern California (2019) and Bachelor’s degree from Inha University (2017) in South Korea, both in Computer Science.

Publications

(Preprint) Symmetric Reinforcement Learning Loss for Robust Learning on Diverse Tasks and Model Scales
Ju-Seung Byun, Andrew Perrault
arxiv 2024 / Paper / Code

(Preprint) Reinforcement Learning for Fine-tuning Text-to-speech Diffusion Models
Jingyi Chen, Ju-Seung Byun, Micha Elsner, Andrew Perrault
arxiv 2024 / Paper

ARES: Alternating Reinforcement Learning and Supervised Fine-Tuning for Enhanced Multi-Modal Chain-of-Thought Reasoning Through Diverse AI Feedback
Ju-Seung Byun*, Jiyun Chun*, Jihyng Kil, Andrew Perrault
EMNLP 2024 / Paper / Code

SalsaBot: Towards a Robust and Generalizable Embodied Agent
Chan Hee Song*, Jiaman Wu* Ju-Seung Byun, Zexin Xu, Vardaan Pahuja, Goonmeet Bajaj, Samuel Stevens, Ziru Chen, Yu Su
Embodied AI Workshop at CVPR 2023 / Paper (Short)
Alexa Prize SimBot Challenge Proceedings 2023 / Paper (Long)

(Preprint) Normality-Guided Distributional Reinforcement Learning for Continuous Control
Ju-Seung Byun, Andrew Perrault
arXiv 2023 / Paper / Code

Training Transition Policies via Distribution Matching for Complex Tasks
Ju-Seung Byun, Andrew Perrault
ICLR 2022 / Paper / Code / Video

Proximal Policy Gradient: PPO with Policy Gradient
Ju-Seung Byun, Byungmoon Kim, Huamin Wang
arXiv 2020 / Paper / Code