Artificial Intelligence

AI Past Courses by J.-L. Liu

工業革命：動力1.0 (1760)，電力2.0 (1870)，數位3.0 (1945)，智力4.0 (2016 AlphaGo)

AI 聽說讀寫食衣住行育樂醫金…, ABC.

l AI abc: An Introduction to Machine Learning

l Gradient Descent and Backpropagation in Machine Learning (Automatic Differentiation: Forward & Reverse Modes, Jacobian)

l Convolution in Machine Learning (Convolution)

Part I Computer Programming (Browse and Use) GitHub

1. Colab: Proj1: tf3-MNIST.py, keras_tensorboard.ipynb (Tutorial)
A. Style Transfer (YouTube2, Code2), B. Time Series (YouTube3), C. YOLO (YouTube4, Code3), D. Stock (YouTube5, Code4), E. Game (OpenAI RL Code5)

2. TensorFlow, TensorBoard (YouTube1)

3. PyTorch (PyTorch Autograd, PyTorch入門, MNIST.ipynb, MNIST.py)

4. Python Programming

5. C++ Programming

A. Cloud Computing by Colab: Google Chrome => Google Account => Google 雲端硬碟 => Click tf1.ipynb => Upload tf1.ipynb to Google 雲端硬碟 => Click on tf1.ipynb => Install Colab Notebooks => Run tf1.ipynb (Done!)

B. Local Computing by jupyter: Install Anaconda3-4.2.0 (or more Anaconda) => Anaconda Navigator => Environments => Install TensorFlow => Home => Launch jupyter => jupyter => Files on 筆電 => Click on tf1.ipynb => Run tf1.ipynb (Done!)
How to run py code on jupyter: tf1.py => Creat a new file tf1.ipynb with only one line “import tf1” => Run the cell of “import tf1” (Done!)

C. Install PyTorch via Anaconda: Click Anaconda Prompt => >conda install PyTorch -c PyTorch => (take a while) => >pip install torchvision => Try 1. import numpy as np 2. import torch 3. x = torch.empty(5, 3) 4. print(x) on jupyter (Done!)

D. Local Computing by PyCharm:

1. tf1.py: # Put this file tf1.py in D:\AI\TF (in English, no Chinese). 
# Install Anaconda3-4.2.0 => Install PyCharm Community Edition 2017.1.5 for Windows 
=> PyCharm => Create New Project => Location: D:\AI\TF 
=> Interpreter: Click on Create Conda Env (Far Right Button) 
=> Click Create => Click File => Open tf1.py => Tools 
=> Python Console => >>> import pip => >>> pip.main(['install', 'tensorflow']) 
=> Wait for >>> => Click the green triangle button (Run tf1). Done!
# See YouTube Link for more if needed.
2. tf2.py: # How to use TensorBoard. # PyCharm => View => Tool Windows 
=> Terminal => Click the green + sign in the Terminal Screen 
=> (C:\Users\Jinn\Anaconda3\envs\untitled1) D:\AI\TF> tensorboard --logdir="./graphs" --port 6006 
=> Copy http://Jinn-PC:6006 => Google Chrome 
=> Paste http://Jinn-PC:6006 to Chrome's http address 
=> GRAPHS (the graph of a, b, Add)
3. TF mnist 1.0 (tf3-MNIST.py) => # Google Chrome => http://Jinn-PC:6006 => GRAPHS => # Error: matplotlib not installed 
=> # PyCharm => File => Setting => Project: TF => Project Interpreter => + => matplotlib

=> Specify version => Install # Save zout_MNIST_1.0.png

Part II Supervised Learning (Read and Work)

1. A Simple Learning Model: Classification, Target, Hypothesis, Training Data, Learning Algorithm, Weights, Bias, Supervised and Unsupervised Learning

2. Google Tutorial for ML Beginners: Image Recognition, MNIST, Softmax Regression (92%), Cross Entropy, Gradient Descent, Back Propagation, Computation Graph (TF mnist 1.0)

3. Tensorflow and Deep Learning I (by Martin Gorner) : Deep Learning Network, ReLU, Learning Rate (98%), Overfitting, Dropout (98.2%), Convolutional Neural Network (99.3%) (TF mnist 3.1)

4. Tensorflow and Deep Learning II (by Martin Gorner) (RNN1): Batch Normalization (99.5%) (TF mnist 4.2), Data Whitening, Fully Connected Network, TensorFlow API, MNIST Record (Kaggle: 100%), Recurrent Neural Network, Deep RNN, Long Short Term Memory, Gated Recurrent Unit, Language Model

Part III Theories of Deep Learning

1. Lectures at MIT (Book: Deep Learning by Goodfellow, Bengio, Courville)

2. Lectures at Chicago

3. Lectures at Stanford

Part IV Reinforcement Learning (by David Silver, YouTube, Book)

1. Introduction to Reinforcement Learning

2. Markov Decision Processes

3. Planning by Dynamic Programming

4. Model-Free Prediction

5. Model-Free Control

6. Value Function Approximation

7. Policy Gradient Methods

8. Integrating Learning and Planning

9. Exploration and Exploitation

10. Case Study: RL in Classic Games

More AI Links …

[1]* KDD Cup Competition, Data Science Game

[2] Google TensorFlow Frontiers, Non-Experts, ML APIs, Dev Summit 2017

[3] OpenAI, OpenAI Universe

Part V AI Seminar

1. 2017.08.01: Udacity Self-Driving Project (Simulator), Auto1-Robocar-Lidar, Auto2-CNN-KITTI-TORCS, Auto3-CNN-OSM-GSV

2. 2017.08.08: D. Lee, Introduction to Humanoid Robotics

3. 2017.09.11: 陳冠霖, Introduction to Visual Common Sense for Autonomous Driving

4. 2017.09.25: Auto4-DeepQNet-Atari (DQN1), Auto5-DQN-TORCS (Patent, TFCode, DQN2, OpenAI Gym (Paper))

5. 2017.09.25: 劉冠漢, At First Glance for Direct Perception and Simulator TORCS in Autonomous Driving

6. 2017.11.13: 朱沛全, Magenta, AI Experiment, Google, Generate Your Own Sound with NSynth

7. 2017.11.27: 劉昌倫, Continuous control with deep reinforcement learning-TORCS

8. 2017.11.27: D. Lee, Introduction to Caffe Framework (Caffe1, Caffe2, Caffe-Code)

9. 2018.01.08: Auto6-VoxelNet (RPN)

10. 2018.03.02: Auto7-TORCS (GoogLeNet-YouTube, GoogLeNet-Tensorflow)

11. 2018.03.19: Sydney Lin, Google USA, My Experience to and at Google

12. 2018.03.19: 劉冠漢, A very simple example of building new tracks

13. 2018.03.26: D. Lee, Algorithm for distance indicators

14. 2018.04.09: 廖鴻志, Introduction to GoogLeNet

15. 2018.04.16: 陳冠霖, Installation of DeepDriving

16. 2018.04.23: D. Lee, TORCS AI driving data for opponent distances

17. 2018.04.30: 劉昌倫, Deep Learning Algorithm for Autonomous Driving using GoogLeNet

18. 2018.05.14: Auto8: World Models (RL, RNN, VAE, MDN, V, M, C, OpenAI Gym, Car) (Code)

19. 2018.11.02: D. Lee, Tasks and ML approaches of autonomous driving

20. 2018.11.09: 陳冠霖, Implementation of DeepDriving (2018 Open Data 分析競賽)

21. 2018.11.23: 王偉誠, Robot Path Planning with TSP by Solving Pointer Networks

22. 2018.12.21: D. Lee, Development of a new controller in DeepDriving

23. 2018.12.28: 劉昌倫, Using Keras and Deep Deterministic Policy Gradient to Play TORCS

24. 2019.01.04: D. Lee, The new controller and proposed 5 indicators

25. 2019.04.12: Google TossingBot (Website, Paper, PyTorch, CNN, RL, Physics)

News

1. 2021.11.12: 國際自駕車趨勢與挑戰

1. 2020.12.24: 新創公司Nuro取得加州第一張許可

2. 2020.12.23: 蘋果打造自駕車輛傳於2024年量產

3. 2020.12.17: 聯發科成為MLCommons聯盟的創始成員

4. 2020.11.13: 語調激似真人！南韓首位AI主播亮相

5. 2020.06.26: 投資自駕車最大手筆 亞馬遜12億美元收購Zoox

6. 2019.09.17: 威盛用智慧型手機駭進 CRV 變自動駕駛

7. 2019.08.01: 「自駕腳踏車」AI 跨出了超越人類智慧的第一步！

8. 2019.06.14: AWS首度在臺北舉行自動駕駛迷你車比賽

9. 2019.06.12: 交大AI團隊研發自駕車發嵌入式AI物件辨識系統

10. 2019.05.21: 2019全國智慧製造大數據分析競賽 (總獎金220萬)

11. 2019.04.24: 國家隊自駕科研成果吸睛 工研院自駕車南寮上路！

12. 2019.03.29: Google機器人TossingBot

13. 2019.03.28: 從AI三大趨勢看台灣廠商的發展策略

14. 2019.02.19: 2019中華郵政大數據競賽 (總獎金100萬)

15. 2019.02.19: Google、亞馬遜、臉書投入AI晶片研發

16. 2019.02-18: Another AI Winter?

17. 2018.12.05: AI幫醫師癌症治療決策下得更快

18. 2018.11.09: 教育部智慧聯網專題實務競賽

19. 2018.10.02: 2018 Open Data 分析競賽

20. 2018.09.20: 全球人工智慧新創的募資概況

21. 2018.09.18: 日盛黑客松競賽 (總獎金25萬)

22. 2018.08.31: 自駕車的未來與挑戰 胡竹生、王傑智（交大電機系與工研院機械所）

23. 2018.07.17: 教育部全國智慧製造大數據分析競賽 (總獎金220萬)

24. 2018.05.32: Where should I start learning AI?

25. 2018.05.16: Data Science Game 2018 (法國)

26. 2018.05.12: Running Robot, Robot Dog

27. 2018.04.26: 科技部創新創業激勵計畫

28. 2018.04.25: 科技部AI創新研發成果展

29. 2018.03.02: 20 Math Books for Machine Learning

30. 2018.02.02: 鴻海搶攻AI 五年內砸100億

31. 2017.12.31: Deep Learning Achievements of 2017 (Part 2)

32. 2017.12.31: Deep Learning Achievements of 2017 (Part 1)

33. 2017.12.05: How can I learn the big picture of Machine Learning?

34. 2017.11.27: 聯發科廣招AI人才 投資平民自駕車AutoX (AutoX)

35. 2017.11.19: 英政府力推無人車相關產業 希望2021上路

36. 2017.10.28: What is deep reinforcement learning?

37. 2017.10.26: Can AI find prime numbers?

38. 2017.10.25: AI人才稀缺 科技廠一擲千金

39. 2017.09.18: AI Pioneer Says We Need to Start Over

40. 2017.06.14: 蘋果開發自動駕駛系統，那是「所有AI之母」

41. 2016.08.16: Is deep learning overhyped?

Part VI Self-Driving Cars

1. Introduction to Self-Driving Cars

l Carnegie Mellon U 1989, CMU Vehicle, Computer

l comma.ai openpilot 2018, commaai, comma-GitHub, openpilot

l Tesla Autopilot 2019, auto vs open

l Self-Driving Car, Autonomous Car

2. End-to-End Learning for Autonomous Driving

l Video, Poster, *Paper

3. Project 1: Steering Angle

l *comma coding

l Toyota Dynamic Radar Cruise Control, Adaptive Cruise Control,

4. Project 2: Lane Detection

l Code 2, Data 2, TuSimple

l Toyota Lane Tracing Assist, Lane Centering

5. Project 3: Speed Prediction

l Code 3, Data 3, TuSimple

6. Project 4: Localization

l Code 4, Data 4, Laika

l Theory: GNSS Processing, Trilateration, Least Squares,

7. Driving Video Dataset

l comma Data

l Udacity Data

l Nvidia Data

l Berkeley Data, GitHub

8. Hardware and Software in Self-Driving Cars

9. Longitudinal and Lateral Control

10. CAN Bus Protocol

11. Environment Perception

12. Traffic Signs Detection

13. Pedestrian Detection

14. How to ensure the safety of Self-Driving Cars

comma Coding

Introduction: comma.ai, comma two (CTwo), *openpilot (OPGit, OP1, OP2, OP3, OP4)

Machine Learning: AIabc, CNN, RNN

NN: Keras, Krs1, Krs2, Krs3, Krs4, 10CNNs, ResNet, OPNet, RetinaNet, Yolact, Demo, Zoox

Homework
HW1: Run and read (A) JL1.py (OPNet, Leon, LeonB, Shen) (B) TT1.py (Yola20, IS1, IS2, Yolo18, RetN17, SSD15, OF13)
HW2: Run and read mnist.py (py2pb), mobile.py, retina.py
HW3: Install Ubuntu 20.04 by VMware (Guide1, Guide2, Guide3)
HW4: Install and run openpilot (UI). Read unlogger.py, ui.py.
HW5: Read view_steering_model.py

HW6: Read train_steering_model.py and server.py

Projects
Project 1: Steering and Lane Detection: HW1A, HW5, HW6
Project 2: Instance Segmentation: HW1B, HW2, SNPE (pb2dlc)
Project 3: Deployment: HW1A, Shen, SNPE, CTwo, WB1
Project 4: Control: Project 3, OP2, Tune, Panda (PD1), CAN

2017 Summer Course

人工智慧程式設計Artificial Intelligence Programming

A Short Scientific Programming Course for All Students

Prerequisites: Undergraduate Calculus and Programming

Time: 1 ~ 4pm, Tuesdays, July 4 ~ Aug. 8 (6 weeks), 2017

Place: 校本部綜三館203

Please register online 線上註冊 for this course.

Instructor: Jinn-Liang Liu 2017.5.21

2017 Fall Course

人工智慧程式設計Artificial Intelligence Programming

Lecture: PHYS物 504, 1:20pm, Mondays (2017.9.11 – 12.25)

Seminar: PHYS物 504, 2:45pm, Mondays (2017.9.11 – 12.25)

Grading: Coding Projects and Reports 100%

Reports: Send reports in ppt (short) and pdf (long) files to jinnliu@mail.nd.nthu.edu.tw by 10/15; 11/15; 12/25.

2018 Winter Seminar

人工智慧專題Topics in Intelligence Programming

Seminar: GEN II綜二402, 2pm, Mondays (Jan 8, Jan 22, Feb 5)

2018 Spring Course

人工智慧專題 Topics in Artificial Intelligence

Seminar: PHYS物 501, 1:20 - 4pm, Mondays (2018.2.26 – 6.11)

Grading: Coding Projects and Reports 100%

Reports: Send reports in ppt (short) and pdf (long) files to jinnliu@mail.nd.nthu.edu.tw by 3/12; 4/2; 4/22; 5/13; 6/3.

2018 Summer Short Course

機器學習簡介 Introduction to Machine Learning

A Short ML Programming Course for All Students

Prerequisites: Undergraduate Calculus and Programming

Time: 10am ~ 12, 1:30 ~ 4:30, July 16 (Monday) & July 19 (Thursday), 2018

Place: GEN III綜三館837

Please register online 線上註冊 for this course.

Course Info: Google Jinn-Liang Liu for details. 2018.5.25

2018 Summer Seminar

人工智慧專題 Topics in Artificial Intelligence

Seminar: GEN II綜二計科所402 (非學科所402), 2:30pm on June 11, 25; July 9, 23, 30; Aug. 27; Sept. 3, 7.

2018 Fall Course

機器學習 Machine Learning

Lecture: ENG1工一 217, 1:20 – 1:35pm, Fridays (2018.9.14 – 2019.1.4)

Seminar: ENG1工一 217, 2:50 – 4:05pm, Fridays (2018.9.14 –2019.1.4)

Reports: Do Proj1, 2, 3, 4 and Send reports in ppt (short) and pdf (long) to jinnliu@mail.nd.nthu.edu.tw by Wednesdays 10/10; 11/14; 12/26.

2019 Winter Seminar

人工智慧專題 Topics in Artificial Intelligence

Seminar: GEN II綜二計科所, 3:00-5:00, Wednesdays (2019.1.9 (綜二A402), 1.16 (綜二A813), 1.23 (綜二A813)).

2019 Spring Course

深度學習 Deep Learning

Lecture: ENG1工一 217, 1:00 – 2:50pm, Fridays (2019.2.22 – 2019.6.14)

Seminar: ENG1工一 217, 3:00 – 3:50pm, Fridays (2019.2.22 – 2019.6.14)

Reports: Do Proj1 and a project of your choice and Send reports in ppt (short) (Demo) and pdf (long) to jinnliu@mail.nd.nthu.edu.tw (email 標題：姓名 學號 Report 1, 2, or 3) by Wednesdays 3/27; 5/1; 6/12 (and 5-Minute Presentation in mp4). 5-Minute Presentations (Demo) on 5/24; 5/31. Demos and Discussions 6/14.

2019 Fall Course

深度強化學習 Deep Reinforcement Learning

Lecture and Seminar: ENG1工一 217, 1:20 – 4:10pm, Thursdays (2019.9.12 – 2020.1.2)

Reports: Send reports in ppt (short) (Demo) and pdf (long) to jinnliu@mail.nd.nthu.edu.tw (email 標題：姓名 學號 Report 1, 2, or 3). Report 1 on Proj1, due 10/3. Report 2 on Your Project, due 11/14. Report 3 on Your Project AND Presentation File mp4, due 12/19. 5-Minute Presentations (Demo) on 12/26 and 1/2.

2020 Spring Course

Topics in Machine Learning 機器學習專題

Online Course Meetings on MS Teams. How to join? Check your email sent by Teams.

Online 1:20 – 4:10pm, Wednesdays (2020.4.1 – 5.6)

ENG1工一 209, 1:20 – 4:10pm, Wednesdays (2020.3.4 – 3.25; 5.13-6.17)

Requirements: Laptop, Python, Calculus, Physics, Lots Reading, Q&A

Send reports in long docx (demo) and short ppt (demo) or short mp4 (demo) to jinnliu@mail.nd.nthu.edu.tw (Email Subject: Name, Student ID No., Report 1, 2, or 3.)

Project Reports 1, 2, and 3 are due on 4/8, 5/13, and 6/17, respectively. 15-Minute Presentations (in mp4) on 6/17.

2020 Summer Seminar

Machine Learning for Self-Driving Cars

Online (mostly) Seminar Meetings on MS Teams. How to join? Check your email invited by me and sent by Teams.

Online 2 – 5 pm, Wednesdays (2020.6.24 – 9.2) *comma coding

2020 Fall Course (Seminar)

Machine Learning for Self-Driving Cars

ENG1工一 209, 1:20 – 3:10 (3:20 – 5:20) pm, Wednesdays (2020.9.16 – 2021.1.6)

Send reports to jinnliu@mail.nd.nthu.edu.tw (Email Subject: Name, Student ID No., Report 1, 2, or 3.)

Report 1: ppt (demo) by 10/21. Report 2: ppt by 11/25. Report 3: ppt and docx (demo) by 1/6.

Presentations on 12/30 and 1/6. *comma coding

2021 Winter Seminar

Machine Learning for Self-Driving Cars

校本部綜二館 GEN II 813, 2 – 5 pm, Wednesdays (Jan 20, 27; Feb 3, 2021) *comma coding

2021 Spring Course

Topics in Deep Learning 深度學習專題 *comma coding

Online 2:00 – 5:00pm, Wednesdays (2021.3.3 – 6.16) Check your email sent by Teams.

On Campus GEN II綜二A813, 1:20 – 5:10pm, Thursdays (2021.3.4 – 6.17)

Send reports to jinnliu@mail.nd.nthu.edu.tw (Email Subject: Name, Student ID No., Report 1, 2, or 3.)

Report 1: ppt (demo) by 3/31. Report 2: ppt by 5/5. Report 3: ppt and docx (demo) by 6/16.

2021 Fall Course

Deep Learning 深度學習 *c omma coding (Report)

Online/On Campus Class 1:20 – 4:10 pm, Wednesdays, 10/6-1/5 by Teams/in GEN II綜二A813

Online Class 1:20 – 4:10 pm, Wednesdays, 9/15-9/29 by Teams

Please email me for Teams Link, if you did not register the course.

Send reports to jinnliu@mail.nd.nthu.edu.tw (Email Subject: Name, Student ID No., Report 1, 2, or 3.)

Report 1: ppt (demo) by 10/27. Report 2: ppt by 12/1. Report 3: ppt and docx (demo) by 1/5.

comma Coding

Introduction: comma.ai, comma two (C2), *openpilot (OPGit, OP1, OP2, OP3, OP4)

Machine Learning: AIabc, CNN, RNN

NN: Keras, Krs1, Krs2, Krs3, Krs4, Krs5, 10CNNs, ResNet, Yolact, OPNet* (OPN)

Homework
HW1: Do Install Ubuntu and OP, Step 1, Step 2
HW2: Run and Read modelA1.py, modelB1.py (Demo)

Projects
Project A: Instance Segmentation (Demo, IS, c omma10k, Cityscapes, Apollo)
Project B: Path Prediction (c omma2k19)

AI Web Seminar

4:15 – 5:30 pm, Wednesdays, 2/16 – 6/8, 2022

Welcome to join us!

It is about OP coding on self-driving cars.

Please send me an email for a Microsoft Teams link to join.

AI Web Seminar

2022 Spring Course

Machine Learning 機器學習 *OP Coding

1:20 – 4:10 pm, Wednesdays, 2/16-6/8, GEN II綜二A813 on-campus or online (Teams link sent by email)

Send reports (in ppt, demo) on Project B5 to jlliu@mx.nthu.edu.tw (Email Subject: Name, Student ID No., Report 1, 2, or 3)

Report 1: Due by 3/30. Report 2: 5/4. Report 3: 6/8.

*OP Coding

Lecture Notes

· *AI abc: An Introduction to Machine Learning

· *Gradient Descent and Backpropagation in Machine Learning (Automatic Differentiation: Forward & Reverse Modes, Jacobian)

· *Convolution in Machine Learning (Convolution)

· *Batch Normalization

Part I Supervised Learning

1. A Simple Learning Model: Classification, Target, Hypothesis, Training Data, Learning Algorithm, Weights, Bias, Supervised and Unsupervised Learning

2. Google Tutorial for ML Beginners: Image Recognition, MNIST, Softmax Regression (92%), Cross Entropy, Gradient Descent, Back Propagation, Computation Graph (TF mnist 1.0)

3. *Tensorflow and Deep Learning I (by Martin Gorner): Deep Learning (98%), ReLU, Learning Rate, Overfitting, Dropout (98.2%), Convolutional Neural Network (CNN, 99.3%) (TF mnist 3.1)

4. *Tensorflow and Deep Learning II (by Martin Gorner) (RNN1): Batch Normalization (99.5%) (TF mnist 4.2), MNIST Record (Kaggle: 100%), Recurrent Neural Network, Deep RNN, Long Short Term Memory, Gated Recurrent Network

Part II Self-Driving Cars

1. Introduction to Self-Driving Cars

· Carnegie Mellon U 1989, CMU Vehicle, Computer

· comma.ai openpilot 2018, commaai, comma-GitHub, openpilot

· Tesla Autopilot 2019, auto vs open

· Self-Driving Car, Autonomous Car

2. OP Coding

3. Data: c omma2k19, c omma10k, Cityscapes, Apollo,

4. How to ensure the safety of Self-Driving Cars