題目：Error-free Training for Artificial Neural Networks

報告人：Bo Deng,  內(nèi)布拉斯加大學(xué)林肯分校,  教授

時間:  06月27日（周四）上午10: 30-11: 20

地點(diǎn)：X30456

摘要：Models of Artificial Neural Networks play an essential role in Artificial Intelligence. All ANN models must be trained before they are deployed to perform tasks. The majority of AI training is supervised. For large-scale models, there are no known methods to achieve 100% accuracy for supervised training. In this talk, I will discuss a newly discovered method that can train ANN models to perfect precision. I will outline the ideas from Dynamical Systems that guarantee the convergence of the error-free training algorithm, and show simulations on the most popular benchmark data for training algorithms in the field. I will also discuss the relationship between the ANN training problem and the classification problem of finite points in Euclidean space that is based on the Stone-Weierstrass approximation theorem in Analysis.

個人簡介：鄧波，八一年學(xué)士學(xué)位：復(fù)旦大學(xué)數(shù)學(xué)系七七屆。八七年博士學(xué)位：密執(zhí)安州立大學(xué)應(yīng)用數(shù)學(xué)。博士后：1987-1988，布朗大學(xué)。現(xiàn)任內(nèi)布拉斯加大學(xué)林肯分校數(shù)學(xué)系教授。主要學(xué)術(shù)研究領(lǐng)域：動力系統(tǒng)，生物數(shù)學(xué)，人工智能。

                          主辦：kaiyun開云官方網(wǎng)站研究生院  

                               承辦：kaiyun開云官方網(wǎng)站

                                          kaiyun開云官方網(wǎng)站數(shù)學(xué)中心