課程名稱 |
深度學習實作與應用 Deep learning and its applications |
開課學期 |
110-1 |
授課對象 |
管理學院 資訊管理學研究所 |
授課教師 |
黃意婷 |
課號 |
IM5062 |
課程識別碼 |
725 U3900 |
班次 |
|
學分 |
3.0 |
全/半年 |
半年 |
必/選修 |
選修 |
上課時間 |
星期二2,3,4(9:10~12:10) |
上課地點 |
管一204 |
備註 |
限本系所學生(含輔系、雙修生) 總人數上限:25人 |
|
|
課程簡介影片 |
|
核心能力關聯 |
核心能力與課程規劃關聯圖 |
課程大綱
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
課程概述 |
The class is designed to introduce students to the background knowledge of deep learning and its applications in many fields. Key concepts of building neural networks will be studied, and the implementations will be demonstrated. The first part of the course will cover the basics of neural networks, including perceptron, forward and backward propagation, regularization, and normalization. Moreover, deep learning models, such as convolutional neural networks, recurrent neural networks as well as sequence models involving attention mechanisms will be included in this course. Also, pre-trained models, i.e. word2vec and BERT, will be introduced. Finally, several domain experts will be invited to present their expertise and experience in many different fields, e.g. computer vision, music, and finance. The final project will involve training a neural network and apply it to a real-world problem. After taking the course, students will obtain the skills for building neural networks on practical problems. |
課程目標 |
The class is designed to introduce students to the background knowledge of deep learning and its applications in many fields. Key concepts of building neural networks will be studied, and the implementations will be demonstrated. |
課程要求 |
- Python: You will probably be fine if you have a lot of programming experience but in a different language.
- Basic knowledge of Probability, Calculus and Linear Algebra: You should know basics of probabilities and distributions, etc, and feel comfortable to take derivatives and understand matrix vector operations and notation.
|
預期每週課後學習時數 |
|
Office Hours |
另約時間 備註: by appointment |
指定閱讀 |
|
參考書目 |
- Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. Retrieved from http://www.deeplearningbook.org/
- Quinn, Joanne, Joanne McEachen, Michael Fullan, Mag Gardner, and Max Drummy. Dive into deep learning: Tools for engagement. Corwin Press, 2019. Retrieved from https://d2l.ai/
|
評量方式 (僅供參考) |
No. |
項目 |
百分比 |
說明 |
1. |
Assignments |
40% |
|
2. |
Midterm exam |
30% |
|
3. |
Final project |
30% |
|
|
週次 |
日期 |
單元主題 |
第1週 |
2021/9/28 |
Course introduction |
第2週 |
2021/10/5 |
Basic Neural Network (I): from regression to neural networks. |
第3週 |
2021/10/12 |
Basic Neural Network (II): backward propagation. |
第4週 |
2021/10/19 |
Basic Neural Network (III) |
第5週 |
2021/10/26 |
Convolutional Networks |
第6週 |
2021/11/2 |
Recurrent Neural Networks |
第7週 |
2021/11/9 |
Word Vector Representations |
第8週 |
2021/11/16 |
Midterm |
第9週 |
2021/11/23 |
Guest Lecturer (1): Computer Vision |
第10週 |
2021/11/30 |
Sequence to sequence learning, attention mechanism |
第11週 |
2021/12/7 |
Transformer, BERT |
第12週 |
2021/12/14 |
Project proposal. Guest Lecturer (2): Music |
第13週 |
2021/12/21 |
Project proposal. Guest Lecturer (3): Finance |
第14週 |
2021/12/28 |
Guest Lecturer (4) |
第15週 |
2021/1/4 |
Practical methodology |
第16週 |
2021/1/11 |
Project presentation |
|