Papers
1
Total Citations
2
H-Index
1
About
JaeUk Kim is a researcher at the forefront of efficient machine learning and multi-agent systems, with a primary focus on real-time sparse training and its hardware acceleration. His most notable contribution is the development of "LearningGroup," a novel framework that enables real-time sparse training on FPGA through learnable weight grouping for multi-agent reinforcement learning (MARL). This work addresses a critical bottleneck in deploying MARL—a powerful technology for applications like multi-robot control and autonomous driving—by dramatically reducing computational overhead without sacrificing model performance. By integrating sparsity directly into the training pipeline, Kim’s approach allows complex multi-agent systems to learn and adapt in real-time on resource-constrained hardware. While his citation count is still growing, the practical implications of his work for edge computing and embedded AI are significant. Kim’s research bridges the gap between algorithmic innovation and hardware implementation, making him a promising figure in the push toward efficient, deployable artificial intelligence for interactive, multi-agent environments.
Research Focus
Key Achievements
Top Papers
- 1