Won-Seok Choi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Won-Seok Choi has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Byoung-Tak Zhang
Seoul National University
2 shared publications - Hyeonseo Yang
Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea
2 shared publications Dong-Sig Han
1 shared publication- Junseok Park
LG AI Research, Seoul, Republic of Korea
1 shared publication - Min Whoo Lee
Department of Computer Science and Engineering, Seoul National University, Seoul, Republic of Korea
1 shared publication - Minsu Lee
School of AI Convergence, Sungshin Women’s University, Seoul 02844, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 10 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 700
Towards object-centric spatial abstraction for probabilistic environment understanding
- 10 July 2026
- Published by Elsevier in Neurocomputing
- Vol. 700
Open AccessTowards object-centric spatial abstraction for probabilistic environment understanding
- Open Access
- 11 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive and Developmental Systems
- Vol. 18 (3) , 637-654
From Sparse to Dense: Toddler-Inspired Reward Transition in Goal-Oriented Reinforcement Learning
Reinforcement learning (RL) agents face fundamental challenges in balancing exploration and exploitation, particularly when sparse or dense rewards bias learning toward sub–optimal behaviors. Biological systems, notably human toddlers, naturally navigate this balance through developmental transitions from free exploration to goal-directed behavior. Inspired by this progression, we propose the Sparse-to-Dense (S2D) reward transition for goal-oriented RL tasks. Our approach transitions from sparse rewards encouraging broad exploration to dense rewards guiding efficient exploitation, while preserving optimal policies via potential-based reward shaping throughout. Through extensive experiments on robotic manipulation and egocentric 3D navigation tasks, we demonstrate that S2D transitions significantly enhance sample efficiency, and generalization compared to static reward schemes and intrinsic motivation baselines. Using a novel Cross-Density Visualizer, we reveal that S2D transitions smooth the policy loss landscape, resulting in wider minima associated with improved generalization. Furthermore, reinterpreting Tolman’s maze experiments in modern RL contexts, we show that early sparse-reward exploration establishes robust initial parameters analogous to cognitive maps, facilitating stable learning under subsequent dense rewards. Our findings bridge developmental psychology and machine learning, offering a principled framework for designing adaptive reward structures in complex RL environments.
- 11 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive and Developmental Systems
- Vol. 18 (3) , 637-654
Open AccessFrom Sparse to Dense: Toddler-Inspired Reward Transition in Goal-Oriented Reinforcement Learning
Reinforcement learning (RL) agents face fundamental challenges in balancing exploration and exploitation, particularly when sparse or dense rewards bias learning toward sub–optimal behaviors. Biological systems, notably human toddlers, naturally navigate this balance through developmental transitions from free exploration to goal-directed behavior. Inspired by this progression, we propose the Sparse-to-Dense (S2D) reward transition for goal-oriented RL tasks. Our approach transitions from sparse rewards encouraging broad exploration to dense rewards guiding efficient exploitation, while preserving optimal policies via potential-based reward shaping throughout. Through extensive experiments on robotic manipulation and egocentric 3D navigation tasks, we demonstrate that S2D transitions significantly enhance sample efficiency, and generalization compared to static reward schemes and intrinsic motivation baselines. Using a novel Cross-Density Visualizer, we reveal that S2D transitions smooth the policy loss landscape, resulting in wider minima associated with improved generalization. Furthermore, reinterpreting Tolman’s maze experiments in modern RL contexts, we show that early sparse-reward exploration establishes robust initial parameters analogous to cognitive maps, facilitating stable learning under subsequent dense rewards. Our findings bridge developmental psychology and machine learning, offering a principled framework for designing adaptive reward structures in complex RL environments.