Survey of Intelligent Technologies 2024

Fall 2024, Friday. 9:10am to 12:00pm, Classroom: B1008 (Management building 10th floor)
Instructors: Yu-Chung Wang, Lun-Wei Ku, and Chih-Yuan Yang

Course Information

In this course, we will introduce three fundamental topics in artificial intelligent computer vision: deep learning (DL), natural language processing (NLP), and computer vision (CV). We expect students to get the concepts of the three fields through our lectures if they are experts in other academic fields but have little background training experience in AI. Each of DL, NLP, and CV will be lectured in five to six weeks, and the overall grade of this course is averaged by a student’s performance in the three divisions.

Prerequisites

• Sufficient ability in English listening, speaking, reading, and writing.
• Experience in programming languages, particularly Python, the widely used language for intelligent computation.
• Knowledge of college-level mathematics.

Syllabus

Week	Date	Topic	Slides	Recording	Action
1	9/6	(DL) Quick review of required math
2	9/13	(DL) Quick review of PyTorch
3	9/20	(DL) Computation graph
4	9/27	(DL) Basic elements in DL
5	10/4	(DL) Basic networks
6	10/11	(NLP) Introduction to NLP
7	10/18	(NLP) Classic NLP
8	10/25	(NLP) Commonly used NLP DNN models
9	11/1	(CV) Introduction to computer vision	pptx	YouTube
10	11/8	(NLP) Introduction to Transformer
11	11/15	(NLP) Introduction to large language models (LLMs)
12	11/22	(CV) Classification	pptx	YouTube
13	11/29	(CV) Image Generation	pptx	YouTube
14	12/6	(NLP) NLP applications
15	12/13	(CV) Detection, Segmentation			Presentation slides due
16	12/20	(CV) Session presentation

Detailed Content

(DL Week1) Quick review of required math: Represent data in tensor, matrix multiplication, vector/matrix derivation, Linear Regression, Linear Regression in Gradient Descent, Introduction to CGU GPU platform
(DL Week2) Quick review of pytorch: Creating Tensor, Tensor multiplication, Dataset(builtin/custom) and Dataloader, Optimization: SGD/Momentum/Adam, Training loop, Evaluation, Loss functions and Regulation
(DL Week3) Computation graph: How to construct a computational graph, Back propagation, Layer and Model
(DL Week4) Basic elements in DL: MLP, Activation, Convolution, RNN, Attention
(DL Week5) Basic networks: LeNet, AlexNet, VGG, NIN, InceptionNet, ResNet, Transformer(optional), Vision transformer(optional)

Textbooks

• DL: Dive into Deep Learning by Zhang et al. (2023)
• NLP: Speech and Language Processing by Daniel Jurafsky and James H. Martin (2024)
• CV: Computer Vision: Algorithms and Applications by Richard Szeliski (2022). It is available online for free.

References

Professor Yu-Chung Wang’s lecture recordings for the course Deep Learning 2023

Professor Chih-Yuan Yang’s lecture slides and video recordings for Computer Vision 2023 Computer Vision 2024

Grading

The three sessions will have their individual grades, and the overall grade of this course will be the average of the three session grades.

Contact Info and Office Hour

Yu-Chung Wang: wycca1@gmail.com
Office hours: TBD

Lun-Wei Ku: lwku@iis.sinica.edu.tw Prof. Ku has no office hour and can set time for discussion. Please contact her via emails

Chih-Yuan Yang: cyyang@cgu.edu.tw
Office hours: Tue 10:30~11:30 Management Building Room 1416