2. Instructor Information
|
Name |
Lee Sung Gu |
Department |
Dept. of Electrical Eng. |
Email address |
slee@postech.ac.kr
|
Homepage |
http://esa.postech.ac.kr/ |
Office |
ENGINEERING 2-403 |
Office Phone |
279-2236 |
Office Hours |
Mondays and Wednesdays from 11am-noon
|
3. Course Objectives
이 과목은 인공지능 관련 연구 과목입니다. 본인이 하고 싶은 인공지능 관련 연구를 하면서 수업 학점을 받을 수 있는 좋은 기회입니다. 강의 내용은 인공지능 방법 중 실질적인 Neural Network 기반 방법 (CNN, RNN 등) 에 사용되는 원리 이론, 연산 방법/이유, 연산량 및 이러한 연산을 잘 지원할 수 있는 컴퓨터 프로그래밍 방법 및 컴퓨터 구조에 대한 내용으로 구성되어 있습니다. 각 학생은 본인의 연구 주제를 정하고 세 번 (Proposal, Progress, Final) 발표하는 기회가 있을 것입니다. 강의 언어는 영어이지만 한글 요약 및 추가 설명도 있을 것입니다.
(This is a research class where the student can pursue any research work related to Artificial Intelligence while getting course credit. Lectures will focus on practical AI Neural Network methods, their theoretical foundations, the computations involved and the reason for the use of such computations, amount of computational operations used, their programming methods, and the computer architectures necessary to support such computations. The student will have opportunities to present his/her research work 3 times during the class (Proposal, Progress, Final). The course language is English with Korean summaries.)
4. Prerequisites & require
At least one previously completed computer programming class is required for this course.
5. Grading
Programming or Paper Project - 50%
(Proposal Presentation 10%, Progress Presentation 10%, Final Presentation 10%, Final Report 20%)
Homework Assignments - 20%
Final Exam - 30%
6. Course Materials
Title |
Author |
Publisher |
Publication Year/Edition |
ISBN |
Michael A. Nielsen, Neural Networks and Deep Learning, Determination Press, 2015. Updated version available at http://neuralnetworksanddeeplearning.com
|
|
|
0000
|
|
7. Course References
Online resources.
8. Course Plan
0. Course Introduction
1. Introduction to Neural Networks
2. Python
3. MNIST
4. Backpropagation
5. Main Neural Network Methods
6. Deep Neural Networks
7. Neural Accelerator Architectures
8. Practical Problems and Their Neural Network Solutions
9. Current Research Work in AI/DL
9. Course Operation
1. Lectures
2. Homework Assignments
3. Student Presentations (projects related to solving practical problems are particularly encouraged)
4. Final Exam
10. How to Teach & Remark
11. Supports for Students with a Disability
- Taking Course: interpreting services (for hearing impairment), Mobility and preferential seating assistances (for developmental disability), Note taking(for all kinds of disabilities) and etc.
- Taking Exam: Extended exam period (for all kinds of disabilities, if needed), Magnified exam papers (for sight disability), and etc.
- Please contact Center for Students with Disabilities (279-2434) for additional assistance