kjw9899/icmlab_lecture

Deep Learning Lecture

ICMLab Deep Learning Lecture

안녕하세요. 단국대학교 전자전기공학부 ICMLab 김재원입니다. 딥러닝을 공부하면서 '공부하는 순서'와 '결국 왜 그렇게 되지?'라는 의문이 항상 남아 있었습니다. 연구실 연구원분들과 딥러닝을 처음 공부하시는 분들께서 느낄 수 있는 의문을 해소하고자 직접 강의 영상과 실습을 준비하였습니다.

Hello, I'm Kim Jae-won of ICMLab, Department of Electronic Electrical Engineering, Dankook University. While studying deep learning, the questions of "the order of study" and "why does it end up like that?" always remained. To solve questions that may be felt by researchers in the laboratory and those who are studying deep learning for the first time, we have prepared lecture videos and practices.

Contact to [email protected]

Installation Process

• numpy
• matplotlib
• pandas
• tqdm
• opencv
• pytorch
• hashlib
• json
• os argparse
• Seaborn
• torchvision
• argparse
• tensorflow

Getting Started

모든 강의 영상은 Youtube에 업로드 해놓았습니다.

All lecture videos have been uploaded on YouTube.

Youtube : https://www.youtube.com/channel/UCWWrL41NmOpmyjhsalMqe4A

Contents

Lecture 1

Lec 1.1

• About AI?

• Basic Regression

Lec 1.2

• Probability and Statistics

Lecture 2

Lec 2.1

• Backpropagation

Lec 2.2

• Overfitting

• Dataset (Train, Validation, Test)

• Batch & Epoch

Lecture 3

Lec 3.1

• Regularigation

• Drop out

Lec 3.2

• Gradient Vanishing

• Activation Function

Lec 3.3

• Weight Initialization

Lecture 4

Lec 4.1

• BatchNormalization

Lec 4.2

• Optimizer

Lecture 5

• Convolutional Neural Network (CNN)

Lecture 6

• Advanced CNN

Reference

본 강의는 여러 논문, 강의 영상을 참고하여 제작하였습니다.