讲座标题(中文):
过参数化浅层神经网络的非空泛化界限
讲座标题(英文):
Non-vacuous Generalization Bounds for Overparameterized Shallow Neural Networks
讲座摘要:
Overparameterized neural networks often exhibit a benign overfitting phenomenon, achieving excellent generalization performance despite having more parameters than training samples. Traditional generalization analyses typically yield vacuous bounds due to overparameterization. In this talk, we show non-vacuous generalization bounds by controlling the Rademacher complexity of overparameterized shallow neural networks (SNNs), supported by empirical studies involving highly overparameterized SNNs. Our complexity bounds are fully dependent on the distance from the initialization point and are expressed in terms of the path-norm of the networks.
讲者信息:
雷云文于武汉大学获得博士学位。现任香港大学数学系助理教授。他的研究方向包括机器学习、数据科学、学习理论与随机优化。同时,他担任《机器学习》《机器学习研究汇刊 TMLR》《IEEE 神经网络与学习系统汇刊》副编辑,并兼任国际机器学习大会(ICML)、神经信息处理系统大会(NeurIPS)、国际学习表征会议(ICLR)及国际人工智能与统计会议(AISTATS)的领域主席。