Seoul National University
M1522.001600 Topics in Big Data Analytics
Large Scale Data Analysis Using Deep Learning
Spring 2017 - U Kang

News and Announcements

Course Information

Deep learning is a branch of machine learning based on a set of algorithms that attempt to model high level abstractions in data. Deep learning is a driving force of the recent advances in AI. In this course, we study core techniques of deep learning to analyze large amount of data. Topics include machine learning basics, deep feedfrward networks, regularization, optimization, convolutional networks, recurrent neural networks, etc.


Date Topic Notice
Mar 6 Introduction to this Course
8 Introduction (Chapter 1)
13 Linear Algebra (Chapter 2)
15 Probability and Information Theory (Chapter 3)
20 Probability and Information Theory (Chapter 3): use the previous slide
22 Numerical Computation (Chapter 4)
27 Machine Learning Basics (Chapter 5)
29 Machine Learning Basics 2 (Chapter 5)
Apr. 3 Project proposal presentation
5 Deep Feedforward Networks (Chapter 6)
10 Deep Feedforward Networks 2 (Chapter 6)
12 Deep Feedforward Networks 2 (Chapter 6): use the previous slide
17 Introduction to TensorFlow
19 Midterm
24 Convolutional Networks (Chapter 9)
26 Convolutional Networks 2 (Chapter 9)
May 1 Sequence Modeling: Recurrent and Recursive Nets (Chapter 10)
3 Sequence Modeling: Recurrent and Recursive Nets (Chapter 10): use the previous slide
8 Project progress presentation
10 Sequence Modeling: Recurrent and Recursive Nets (Chapter 10): use the previous slide
15 Regularization for Deep Learning (Chapter 7)
17 Regularization for Deep Learning 2 (Chapter 7)
22 Optimization for Training Deep Models (Chapter 8)
24 Optimization for Training Deep Models 2 (Chapter 8)
29 Practical Methodology (Chapter 11)
31 Applications (Chapter 12)
June 5 Autoencoders (Chapter 14)
7 Project final presentation 1
12 Project final presentation 2
14 Final


Late policy - for all deliverables:


The text book is
Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville. (available in online)


We expect you to have an undergraduate-level knowledge on the following topics: We provide some background, but the class will be fast paced.
Last modified Apr. 3, 2017, by U Kang