深度學習
Deep Learning
學期
108-1
學分
3
學分
當期課號
5981
永久課號
IOG5006
開課單位
智慧科學暨綠能學院
授課教師
魏澤人
校區
歸仁
類別
選修
上課時間表
| 節 | 週一 |
|---|---|
5 13:20–14:10 | 深度學習 CM218 3 節連堂 |
6 14:20–15:10 | |
7 15:30–16:20 |
* 根據陽明交大上課時間表所列
概述
This course intended to help students enter the field of deep learning. We will study the fundamental theory of various neural network architectures and building blocks, including convolutional networks, gradient descent based optimizers and math behind them. Then we will explore a few use cases of deep learning, including generative models and reinforcement learning algorithms.
先修科目
Linear algebra, Multivariable Calculus, Programming(mostly in Python)
教學方式
Lectures and various experiments and projects as coursework.
評分方式
100% Coursework(projects, experiments)
週次計畫
| 週次 | 主題 |
|---|---|
| 第 1 週 | Overview and Enviromen Setup |
| 第 2 週 | Math and Machine Learning Basics |
| 第 3 週 | Deep Feedforward Networks |
| 第 4 週 | Regularization and Training |
| 第 5 週 | Covolutional Networks Basic |
| 第 6 週 | Applications and Reviews |
| 第 7 週 | Recurrent Nets |
| 第 8 週 | Midterm Project. In class competition |
| 第 9 週 | Some Modern Network Building Blocks |
| 第 10 週 | Linear Factor Models |
| 第 11 週 | Autoencoders |
| 第 12 週 | Structured Probabilistic Models for Deep Learning |
| 第 13 週 | Deep Generative Models |
| 第 14 週 | Topics and Applications |
| 第 15 週 | Monte Carlo Methods |
| 第 16 週 | Introduction to Deep Reinforcement learning |
| 第 17 週 | Topics and review |
| 第 18 週 | The Final Lecture |
教科書
Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville, MIT Press, 2016 https://www.deeplearningbook.org/
Office Hours
- 地點
- My Office or by reservation
- 時間
- Tuesday 8:00-10:00
- 聯絡方式
- tjw@nctu.edu.tw