【报告摘要】 The brain is a complex dynamical network whose function emerges from the interplay of oscillatory activity, connectivity, and external inputs. In this talk, I present my research that approaches brain dynamics from three complementary angles. First, I discuss how coupled oscillator networks capture key neural phenomena such as phase-amplitude coupling, remote synchronization, and cluster synchronization. Second, I describe our work on representation learning and contextual bandits for sequential decision-making under non-stationarity and delayed rewards—settings motivated by adaptive brain stimulation. Third, I show how vibrational stabilization and controllability analysis can be used to design principled stimulation strategies for steering brain networks toward desired dynamical regimes. I highlight the interplay between these pillars at the intersection of control theory, machine learning, and computational neuroscience.

【报告人简介】秦语真博士，现任荷兰奈梅亨拉德堡德大学(Radboud University)机器学习与神经计算系助理教授。他于2012年和2015年分别在河海大学和武汉大学获得自动化专业学士和硕士学位，于2019年获得荷兰格罗宁根大学系统控制专业博士学位。2020至2023年间，他于美国加州大学河滨担任博士后研究员。他的主要研究方向包括复杂网络控制、非线性系统控制、强化学习，以及控制理论和机器学习在神经调控、闭环BCI中的应用。