9 papers
V-Simba: Unleashing the Architectural Potential of RL in Visual Continuous Control
Donghu Kim, Youngdo Lee, Hojoon Lee +6
Improving sample efficiency remains a core challenge in reinforcement learning (RL), especially in real-world settings like robotics, where data collection is costly. This challeng…
FlashSAC: Fast and Stable Off-Policy Reinforcement Learning for High-Dimensional Robot Control
Donghu Kim, Youngdo Lee, Minho Park +10
Reinforcement learning (RL) is a core approach for robot control when expert demonstrations are unavailable. On-policy methods such as Proximal Policy Optimization (PPO) are widely…
FIRE: Frobenius-Isometry Reinitialization for Balancing the Stability-Plasticity Tradeoff
Isaac Han, Sangyeon Park, Seungwon Oh +3
Deep neural networks trained on nonstationary data must balance stability (i.e., retaining prior knowledge) and plasticity (i.e., adapting to new tasks). Standard reinitialization…
Dynamic Mixture of Experts Against Severe Distribution Shifts
Donghu Kim
The challenge of building neural networks that can continuously learn and adapt to evolving data streams is central to the fields of continual learning (CL) and reinforcement learn…
Building Resource-Constrained Language Agents: A Korean Case Study on Chemical Toxicity Information
Hojun Cho, Donghu Kim, Soyoung Yang +3
Language agents powered by large language models (LLMs) face significant deployment challenges in resource-constrained environments, particularly for specialized domains and less-c…
Hyperspherical Normalization for Scalable Deep Reinforcement Learning
Hojoon Lee, Youngdo Lee, Takuma Seno +3
Scaling up the model size and computation has brought consistent performance improvements in supervised learning. However, this lesson often fails to apply to reinforcement learnin…