[課程] 2020基於遊戲的機器學習入門

Instructor

蘇文鈺 (Alvin W.Y. Su, alvinsu@mail.ncku.edu.tw)

Teaching Assistant

陳麒麟，丁俊哲，戴逸任，陳如琰 RM 65707

Course Materials

1. Introduction to machine learning textbook
2. Gamed Based coding platform
3. Online ppt
4. Reference: Stanford CS229, https://stanford.edu/~shervine/teaching/cs-229.html

Power Point

•  Week2 
• 課程投影片0_環境安裝
•  Week3
•  Week4 - 線上課程演練週
•  Week5 - 清明連假
•  Week6 
• Introduction  (課程影片0) (課程影片1)
• K-means and KNN (課程影片2) (課程影片3--Lab)

• Week7 - 第一次比賽 **
• Week8 - SVM  SVM_Lab（課程影片4）
• Week9 - Machine Learning Basics and Tips (課程影片5)
• Week10 - Neural Network (課程影片7)
• Week11 - 第二次比賽
• Week12 - test of statistical hypothesis (課程影片8)
• Week13 - CNN (課程影片9)
• Week14 - 加強學習(課程影片10)
• Week15 - 加強學習(續)(課程影片11)

===========================================================
**[第一次比賽_注意事項]

1.登入PassCode 請輸入 CSIEML (請參考沒有聲音的影片)

2.請留下一個版本就好，其餘的版本請刪除，以利考試運作。(沒有刪除斟酌扣0.5分)

3.　4/15(三) 12:00 ~ 4/16(四) 09:00 登入上傳

　　4/16(四) 12:00 前公布考試順序

　　4/16(四) 16:00 直播考試結果

4.　Office Hour 4/15(三) 19:00-21:00　65705
5. 示範影片

**[第二次比賽公告]
公告如下

https://hackmd.io/@Dingjunzhe/SkRBjFlIL/%2Fk6bHBJhyRj6_B3NYjW6CtA

**[第三次比賽公告]

考試公告如下
https://hackmd.io/@Dingjunzhe/SkRBjFlIL/%2Fi6GeXYnuTOeuiSfKqw6y3w

===========================================================

Course Arrangement

1st quarter

Purpose: Get familiar with the course requirement, the offline game platform and the online game platform. Collect and annotate the data.

week 1 : Introduction of the course and applications using ML. How this course will proceed this semester. Especially in the COVID-19 period.

week 2 : Setup of Python programming environment. Use of Github.

week 3 : Build and Practice of the offline Game Platform and the online Game Platform.

week 4 : Practice using the online Game Platform. Intro. to basic classifiers

2nd quarter

Purpose: Application of simple machine learning tools to a computer game, called Breakout Clone, also called Arkanoid. Object positions provided.

week 5 : 兒童節、民族掃墓節補假。

Week 6: Introduction and models

Week 7: 1st online contest, New Breakout Clone (Behavior is a little bit different!) (10%)

Week 8: Support Vector Machine/Multi-class clustering/classification

3rd quarter

Purpose: Application of more machine learning tools to New Pong and Racing Cars.

Week 9: 2nd online contest. New Pong (15%)

Week 10: Bagging, Boosting, Adaboost

Week 11: Neural Network

Week 12: 3rd contest. Racing Cars 1. (20%)

Final quarter

Purpose: Application of machine learning tools to Racing Cars 2. Object positions provided.

Week 13: DNN

Week 14: CNN

Week 15: RNN

Week 16: Team Report

Week 17: 端午節(放假)。

Week 18: Final contest (40%) and final report (5%)

Target

1.  Learn basic machine learning algorithms

2.  Use machine learning tools

3.  Understand data and its preprocessing

4.  Learning by doing and gaming.

Grading by Rank

Reports : 2+3+5+5= 15%   

Contest 1: 10%   

Contest 2: 15%   

Contest 3: 20%   

Contest 4: 40%   

Bonus: 10%

Cours Preparation Items
                 All reports in HackMD 
                 All works in Github 
                 All contest results on the online platform 
                 You may need to learn many materials from online courses and/or videos.