RESEARCH INTERESTS
Nonequilibrium Physics and Machine Learning
- Stochastic thermodynamics and (classical) information thermodynamics
- We found thermodynamics uncertainty relations for non-steady states in both continuous- and discrete-time Markov processes and applied it to measurement and feedback control (Maxwell’s demon).[Phys. Rev. Lett. 125, 140602 (2020)]
- We found universal kinetic bound on the static response of a generic nonequilibrium observable to external perturbations in terms of the dynamical activity (or traffic) that quantifies the frequency of stochastic state transitions of a Markov process, named as response kinetic uncertain relation (R-KUR).[Commun. Phys. 8, 62 (2025)]
- Deep learning theory
- We derived analytical formulae of the noise and model fluctuations of the stochastic gradient descent (SGD) algorithm in deep learning with a finite learning rate.[ICML 2021]
- We analyzed analytically the fundamental properties of the minibatch noise in discrete-time SGD.[ICLR 2022 Spotlight]
- We derived the stationary distribution and found a power-law escape rate from a local minimum for minibatch SGD.[ICML 2022 Spotlight]
- Quantum thermodynamics
- We derived a quantum response kinetic uncertainty relation (QR-KUR) for continuously monitored Markovian open quantum systems, showing that the static response precision of a trajectory observable is bounded by the conventional quantum dynamical activity together with a perturbation-induced inter-subspace transition term that disappears in the classical limit and is further constrained by symmetry-sector selection rules.[arXiv:2501.04895]
- We derived finite-frequency fluctuation-response bounds for Markovian open quantum systems in an input-output setting, showing that any detector’s lock-in response precision is limited by the emitted-field quantum Fisher information and, for dissipative coupling modulation, by calibrated signal-channel activity.[arXiv:2605.03340]
- Quantum Maxwell’s demon
- We constructed a new type of quantum information engine that can store useful work cumulatively and transport a quantum particle unidirectionally by harnessing purely quantum fluctuations with the aid of Maxwell’s demon, whose maximum power and maximum transport velocity are well-defined and the optimal operation time is specified. We proposed an improved definition of the efficiency by including all possible energy flow involved in the engine cycle. We discussed possible experimental implementations with existing techniques, especially those for cold atom systems and optical lattices.[Phys. Rev. A 113, 022436 (2026)]
- We experimentally realized Maxwell’s demon using a 62-qubit superconducting quantum processor, being the first experiment using an isolated quantum many-body system. [Phys. Rev. A 109, 062614 (2024)]
