Hyoseok Byun
Papers
1
Total Citations
20
H-Index
1
About
Hyoseok Byun is a researcher at the forefront of optimization methods for artificial neural networks, with a particular focus on enhancing training efficiency through advanced conjugate gradient techniques. His most notable contribution, the "Variable three-term conjugate gradient method for training artificial neural networks" (2022), has already garnered 20 citations, reflecting its immediate impact on the field. This work introduces a novel algorithm that dynamically adjusts search directions and step sizes, significantly improving convergence speed and stability compared to traditional methods—a critical advancement for deep learning practitioners grappling with large-scale models. Byun’s research bridges the gap between classical optimization theory and modern machine learning, offering practical tools for accelerating network training without sacrificing accuracy. His approach stands out for its mathematical rigor and adaptability, making it valuable for both theoretical studies and real-world applications. As a rising voice in computational optimization, Byun continues to explore how variable parameter schemes can reshape neural network training, promising further breakthroughs in efficient AI system development.
Research Focus
Key Achievements
Top Papers
- 1