Yukawa, Masahiro

数理的基盤に立脚した新しい信号処理パラダイムの構築を目指しています． 数理科学で蓄積された知見を活用することで，信号処理の諸問題を見通し良く解決することが目的です． 主に，不動点近似・凸解析を利用した適応信号処理アルゴリズム等の研究を行なっています． 新しい時代を切り拓く信号処理技術に繋げたいと考えます．

Our goal of research is to build a new mathematical paradigm of (online) algorithm that can be used in a wide range of field such as signal processing, machine learning, and data science. The central achievements so far include the developments of efficient online learning algorithms based on fixed-point approximation, convex analysis, and the theory of reproducing kernel. The efficacy of the proposed approaches has been shown in applications to acoustics, communications, and time-series data prediction.