Math497: Deep Learning Algorithms and Analysis

Get started: Course information and preparations:

• beginning

• ending
  

• Course information, requirements and references   

• Course background and introduction

• Python and Pytorch tutorial

Week1: Linear machine learning models

• Machine learning basics, popular data sets
  

• linearly separable sets

• logistic regression models
  

• KL-divergence and cross-entropy

• support vector machine and relation with LR. 

• optimization and gradient descent method

• MNIST and image classification by logistic regression
  

Week2:  Probability and training algorithms

•   Introduction to probability

• Probabilistic derivation of logistic regression models 

• multi-variable calculus and convex functions

• Stochastic gradient descent method and theory

• MNIST: training and generalization accuracy

Week3

• Nonlinearly Seperable

• Weierstrass

• FEM to DNN

• NonPoynomia DNN

• DNN definition

• Minist

Week4

• FEM to Convolution
  

• Convolutional Operation

• Classic CNNs

• D4
  

• D5
  

Week 5

• Data Normalization

• Batch Normalization in DNNs and CNNs

• Building and Training ResNet with Pytorch

Week 6

• MgMet

• Implementation of Multigrid and MgNet algorithms