Papers
1
Total Citations
1
H-Index
1
About
Hojeong Kim is pioneering the intersection of hyperdimensional computing and federated learning for mobile robotics, addressing critical challenges in resource-constrained autonomous systems. Their most-cited work, "Hyperdimensional Computing-Based Federated Learning in Mobile Robots Through Synthetic Oversampling" (2025), introduces a novel framework that replaces traditional deep neural networks with lightweight, privacy-preserving hyperdimensional vectors. This approach dramatically reduces computational demands while mitigating data heterogeneity through synthetic oversampling—a breakthrough for collaborative robot learning in bandwidth-limited environments. With 1 citation already in its first year, this paper signals growing recognition of Kim's innovative methodology. By decoupling machine learning from heavy neural architectures, Kim's research enables energy-efficient, real-time adaptation for swarms of mobile robots, advancing practical deployments in search-and-rescue, environmental monitoring, and industrial automation. Their work directly tackles two fundamental barriers in edge AI: computational frugality and data privacy. As federated learning expands into robotics, Kim's HD computing framework offers a paradigm shift—proving that brain-inspired vector symbolic architectures can outperform conventional deep learning in distributed, low-power settings. This emerging contribution positions Kim as a rising thought leader in neuromorphic computing and decentralized intelligence.
Research Focus
Key Achievements
Top Papers
- 1