Yusuke Miyajima

Yusuke Miyajima is an Assistant Professor at the School of Advanced Science and Engineering, focusing on magnetism, superconductivity, and strongly correlated systems. His research integrates machine learning with combinatorial optimization problems and phase transitions in spin models.

Research Interests: Dr. Miyajima's work bridges computational physics and neuromorphic AI hardware, particularly targeting Berezinskii-Kosterlitz-Thouless transitions and optimization algorithms inspired by biological systems like amoebas. He investigates mathematical models for physical implementation in devices such as analog circuits and spintronics.

Recent Publications: His studies analyze scaling laws in optimization machines, detect multiple phase transitions using machine learning, and propose simplified models for physical realization. These contributions highlight interdisciplinary approaches in condensed matter physics and artificial intelligence.

Projects: In 2024, he advanced amoeba-inspired optimization machines, demonstrating universal performance improvements for large-scale problems. Earlier work (2023) focused on theoretical implementations using domain walls, enhancing device compatibility and solution efficiency.