Students will learn programming techniques related to artificial intelligence, with a focus on machine learning, through exercises.
For this purpose, students will first learn the Python language, which is commonly used in artificial intelligence programming,
from the basics of syntax to object-oriented programming and the use of major libraries. Then, students will learn how to
express concepts and algorithms such as deep learning, reinforcement learning, and search in Python programs.
Understand the mechanism of artificial intelligence through exercises using Python.
- Develop the efficient program by using Python.
- To write and run codes of deep learning as an artificial intellignece programming.
- To write and run codes of reinforcement learning as an artificial intellignece programming.
- To write and run codes of search as an artificial intellignece programming.
| Class schedule | HW assignments (Including preparation and review of the class.) | Amount of Time Required | |
|---|---|---|---|
| 1. | Introduction ・Building Python development environment ・Python syntax |
Preparation: Read Chapter 2 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| 2. | Deep learning 1. ・Classification with neural networks ・Regression of neural networks |
Preparation: Read Chapter 3 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| 3. | Deep learning 2. ・Convolutional neural network image classification ・ResNet (Residual Network) image classification |
Preparation: Read Chapter 3 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| 4. | Reinforcement Learning 1. ・multi-armed bandit problem ・maze game with policy gradient method |
Preparation: Read Chapter 4 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| 5. | Reinforcement learning 2. ・maze game with Sarsa and Q-learning ・CartPole with DQN (deep Q-network) |
Preparation: Read Chapter 4 of the textbook carefully | 100minutes |
| Review class assignments | 90minutes | ||
| Make a report | 190minutes | ||
| 6. | Search 1. ・Tit-tac-toe by minimax method ・Tit-tac-toe by alpha-beta method |
Preparation: Read Chapter 5 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| 7. | Search 2. ・Tit-tac-toe by Atomic Monte Carlo ・Tit-tac-toe by Monte Carlo Tree Search |
Preparation: Read Chapter 5 of the textbook carefully | 95minutes |
| Review class assignments | 95minutes | ||
| Make a report | 190minutes | ||
| Total. | - | - | 2660minutes |
| Reports | Total. | |
|---|---|---|
| 1. | 25% | 25% |
| 2. | 25% | 25% |
| 3. | 25% | 25% |
| 4. | 25% | 25% |
| Total. | 100% | - |
Evaluation will be based on reports.
Students will receive 60 points or more out of 100 points if they understand the fundamentals of how to program in Python, are able to write executable programs for deep learning, reinforcement learning, and search tasks, and appropriately discuss the results.
Students will receive 60 points or more out of 100 points if they understand the fundamentals of how to program in Python, are able to write executable programs for deep learning, reinforcement learning, and search tasks, and appropriately discuss the results.
| ways of feedback | specific contents about "Other" |
|---|---|
| Feedback in the class |
Textbook: "AlphaZero Deep Learning, Reinforcement Learning, and Search Artificial Intelligence Programming" Eiichi Furukawa
Bourne Digital
Students should have taken a course on artificial intelligence to deepen their understanding.
- Course that cultivates an ability for utilizing knowledge
- Course that cultivates a basic problem-solving skills
| Work experience | Work experience and relevance to the course content if applicable |
|---|---|
| N/A | N/A |
Last modified : Sat Sep 09 07:18:02 JST 2023