Physical reservoir computing
Mesoscopic interference, transient dynamics, nonlinear response, temporal memory, and ridge-regression readouts for neuromorphic information processing.
We study how structure, interfaces, defects, and dynamics determine the function of advanced materials—and translate that understanding into devices and intelligent systems.
Mesoscopic interference, transient dynamics, nonlinear response, temporal memory, and ridge-regression readouts for neuromorphic information processing.
Machine learning, experimental design, and process optimization for complex deposition and synthesis landscapes.
ZnO, SnO, BaSnO₃, β-Ga₂O₃, transparent conductors, thin-film transistors, photodetectors, and power-device concepts.
Quantum-to-classical transport crossovers, mesoscopic conduction, electron–phonon interactions, and engineered interfaces.
Materials and interfaces for energy conversion, solid-state transport, and harsh-environment operation.
Multi-modal sensing, machine vision, sorting, and intelligent process control for value-preserving resource recovery.
Our data-guided materials workflow couples pulsed laser deposition, rapid experimental screening, and machine-learning reconstruction of multidimensional process spaces. In published work on Ga-doped ZnO, this approach identified an optimized region where high optical transparency and low electrical resistivity coexist.
The same philosophy—integrating physical experiments with data-driven inference—guides our broader work in autonomous materials discovery and intelligent processing.

Across our projects, we investigate how interfaces, defects, dimensionality, phase stability, and nonequilibrium dynamics can decouple or coordinate electronic, ionic, thermal, optical, and magnetic behavior.
