Yuma Ichikawa

Statistical-mechanical study of deep Boltzmann machine given weight parameters after training by singular value decomposition

K. Hukushima, Yuma Ichikawa, 2022, 2205.01272.

Statistical-mechanical Study of Deep Boltzmann Machine Given Weight Parameters after Training by Singular Value Decomposition

K. Hukushima, Yuma Ichikawa, 2022, Journal of the Physical Society of Japan.

Toward Unlimited Self-Learning Monte Carlo with Annealing Process Using VAE's Implicit Isometricity

Y. Umeda, Yuma Ichikawa, Akira Nakagawa, 2022, ArXiv.

Dataset Size Dependence of Rate-Distortion Curve and Threshold of Posterior Collapse in Linear VAE

Yuma Ichikawa, Koji Hukushima, 2023, ArXiv.

Toward Unlimited Self-Learning MCMC with Parallel Adaptive Annealing

Y. Umeda, Yuma Ichikawa, Akira Nakagawa, 2022, 2211.14024.

Continuous Tensor Relaxation for Finding Diverse Solutions in Combinatorial Optimization Problems

Yuma Ichikawa, Hiroaki Iwashita, 2024, ArXiv.

Controlling Continuous Relaxation for Combinatorial Optimization

Yuma Ichikawa, 2023, ArXiv.