题目：Parallel Neural Web for Errorless Training Accuracy
报告人：邓波
邀请人：黎定仕
时间：2026年5月28日（星期四）16:00-17:00
地点：X30423
摘要：By definition, training an artificial neural network is finding the global minimum of its loss function. The Gradient Descent Tunneling method solves the training problem in theory. In practice, for training problems with large data sizes, the method is very slow, or not always working. In this talk, we introduce a new model architecture for which the training problem can always be solved quickly. The key difference from the conventional deep ANNs lies in that our new architecture is a parallel web of shallow neural networks, which allows parallel training in short amount of time.
个人简介：邓波，八一年学士学位：复旦大学数学系七七届。八七年博士学位：密歇根州立大学应用数学。博士后：1987-1988，布朗大学。现任内布拉斯加大学林肯分校数学系教授。主要学术研究领域：动力系统，生物数学，人工智能。
