[重要公告] 這門課選修人數眾多且學生背景多元，為確保教學品質，這門課在開學後第 3 個禮
<br>拜有
<br>期初考。期初考範圍為課程網站 (https://nthu-datalab.github.io/ml) 上的 02-
<br>04 Math reviews。如果沒有基礎的同學，請在暑假期間自行觀看老師預錄的影片修習，並跟著 
<br>python notebook 練習上手實作。祝大家有個充實的暑假！
<br>
<br>
<br>一、 課程說明 (Course Desc<x>ription) 
<br>This class introduces the concepts and techniques of deep learning. The 
<br>course consists of 
<br>three parts. In the first part, we give a quick introduction of classical 
<br>machine learning and 
<br>review some key concepts required to understand deep learning. In the second 
<br>part, we 
<br>discuss how deep learning differs from classical machine learning and 
<br>explain why it is 
<br>effective in dealing with complex problems such as the image and natural 
<br>language 
<br>processing. Various CNN and RNN models will be covered. In the third part, 
<br>we introduce the 
<br>deep reinforcement learning and its applications.
<br>
<br>This course also gives coding labs. We will use Python 3 as the main 
<br>programming language 
<br>throughout the course. Some machine learning libraries such as Scikit-learn 
<br>and Tensorflow 
<br>will be explained and heavily used.
<br>
<br>Students are required to have proper understanding of Calculus, Linear 
<br>Algebra, Probability 
<br>Theory, and the Python programming language. Although it would be helpful, 
<br>knowledge 
<br>about classical machine learning is NOT required. 
<br>
<br>IMPORTANT: Students will group (3~4 people a group). This class requires 
<br>**each group of 
<br>students to prepare a GPU card** to perform the necessary computing. You can 
<br>follow this 
<br>li<x>nk (http://timdettmers.com/2017/04/09/which-gpu-for-deep-learning/) to 
<br>decide which 
<br>GPU card 
<br>to go for. NO GPU CARD PROVIDED IN THE CLASS.  
<br>
<br>二、 指定用書 (Textbook) 
<br>Ian Goodfellow, Yoshua Bengio, Aaron Courville, Deep Learning, MIT Press, 
<br>2016, ISBN: 
<br>0387848576
<br>
<br>四、 教學方式 (Teaching Method) 
<br>Video Lecture + Labs
<br>
<br>五、	 教學進度 (Syllabus) 
<br>1.	Fundamentals of Machine learning (5 weeks)
<br>2.	Neural Networks (NNs): Design
<br>3.	NN Optimization and Regularization
<br>4.	Computer Vision & CNNs
<br>5.	Natural Language Processing & RNNs
<br>6.	Unsupervised learning
<br>7.	Semi-supervised & transfer learning
<br>8.	Deep reinforcement learning	(2 weeks)
<br>
<br>六、	 成績考核 (Evaluation)
<br>Quiz: 20% 
<br>Assignments: 40%
<br>Competitions/projects: 40%