【短期课程】Mathematical Theory and Applications of Deep Learning

【短期课程】Mathematical Theory and Applications of Deep Learning

Year:    2022

CAM-Net Digest, Vol. 19 (2022), Iss. 14 : p. 7

Abstract

课程日期:2022年8月16日-30日

授课时间:09:00-11:00

课时数:2课时/天,共30课时

ZOOMID:202 216 6027    密码:Zoom2022

授课老师:Prof. Haizhao Yang (University of Maryland,CollegePark)

主办单位:国家天元数学中部中心、湖北国家应用数学中心、武汉大学数学与统计学院

课程信息: 

Description:

Part I: deep learning basics: feed-forward networks; recurrent neural networks; deep reinforcement learning;

Part II: deep learning applications: data-driven recovery of equations; data-driven prediction; solving partial differential equations; inverse problems, operator learning;

Part III: deep learning theory: approximation theory, optimization theory, and generalization theory of deep learning.

Important prerequisites:

Students must review basic numerical linear algebra, differential equations, probability, and optimization by themselves.

Approximate Schedule:

Each lecture is approximately one hour.

Lecture 1: Basic Machine Learning and Feed-forward Networks;

Lecture 2: Recurrent Neural Networks;

Lecture 3 and 4: Deep Reinforcement Learning;

Lecture 5: Data-driven Recovery of Equations;

Lecture 6: Data-driven Prediction;

Lecture 7 and 8: Solving Partial Differential Equations;

Lecture 9: Inverse Problems;

Lecture 10: Operator Learning;

Lecture 11 and 12: Deep Network Approximation;

Lecture 13 and 14: Deep Learning Optimization;

Lecture 15 and 16: Deep Learning Generalization.

【报名流程及联系人】

1、点击此处提交报名申请

2、请报名的同学加入qq群:743690605(仅用于本次课程通知发布、资料分享);

3、报名截止日期为2022年08月14日

联系人:杨老师  电话:027-68788932   Email:  tmcc@whu.edu.cn

You do not have full access to this article.

Already a Subscriber? Sign in as an individual or via your institution

Journal Article Details

Publisher Name:    Global Science Press

Language:    Multiple languages

DOI:    https://doi.org/2022-CAM-20834

CAM-Net Digest, Vol. 19 (2022), Iss. 14 : p. 7

Published online:    2022-01

AMS Subject Headings:    Global Science Press

Copyright:    COPYRIGHT: © Global Science Press

Pages:    1

Keywords: