The University of Tokyo · Graduate School of Engineering / Faculty of Engineering

Research keywordsneural operators・equivariant neural networks・transformer training stability・function-space learning・robust vision models・urban computing・Reynolds networks

This is a public-data preview. Personalized fit, contact angles, and saved workflows require sign-in.

• QuadNorm: Resolution-Robust Normalization for Neural Operators2026
• Residual Koopman Spectral Profiling for Predicting and Preventing Transformer Training Instability2026
• 手書き漢字の埋め込み表現による反復学習頻度最適化のための半減期予測2025
• Unlocking Noise-Resistant Vision: Key Architectural Secrets for Robust Models2025

• 自治体業務改善に向けたデータ効率型都市点検技術の開発と実応用

• Official profile