Yuto Inui

I am Yuto Inui, a Ph.D. student at The University of Osaka, supervised by Yoshinobu Kawahara.

I study recurrent systems in machine learning from a dynamical-systems perspective, focusing on their long-term behavior. My work explores model design and representation learning in neural differential equations, oscillatory recurrent neural networks, and latent reasoning. I am also interested in generative models, including flow matching and diffusion models.

Research Keywords

• Neural Differential Equations
• Oscillatory Neural Networks
• Perron Frobenius Operator
• Flow Matching