Ho-Taek Joo

About

Ho-Taek Joo is a rising researcher in artificial intelligence, with a primary focus on offline reinforcement learning (RL) and data augmentation techniques. His most-cited work, “A Swapping Target Q-Value Technique for Data Augmentation in Offline Reinforcement Learning” (2022), addresses a critical challenge in applying RL to real-world domains such as healthcare, autonomous vehicles, and robotics, where only fixed, logged datasets are available. Joo’s key contribution is a novel swapping target Q-value method that enhances policy learning by effectively augmenting limited data, overcoming the distributional shift problem that plagues offline RL. With 4 citations, this paper has already garnered attention for its practical approach to improving RL in data-scarce environments. Joo’s work stands out for its direct applicability to safety-critical systems, where collecting new data is costly or risky. As a researcher, he bridges theoretical advances in RL with real-world deployment needs, making his contributions valuable for both academics and practitioners seeking robust, data-efficient learning algorithms.

Research Focus

Key Achievements

Top Papers

1. 1
   A Swapping Target Q-Value Technique for Data Augmentation in Offline Reinforcement Learning

   4 citations · 2022

Key Collaborators

Contact & Links