Course Syllabus

Course Information：

Year-Semester Course Number Course Name Phase Credits Hours Required/Elective Instructor Class Student NO. Withdraw Note

Spring Semester
Year 2025 340860 Deep Learning with TensorFlow 1 3.0 3 ★ Ku, Chiung-Shen 4EE2A
4EE2B
4EE2C 40 0 電機二甲乙丙合開,預選40人

Year-Semester	Course Number	Course Name	Phase	Credits	Hours	Required/Elective	Instructor	Class	Student NO.	Withdraw	Note
Spring Semester Year 2025	340860	Deep Learning with TensorFlow	1	3.0	3	★	Ku, Chiung-Shen	4EE2A 4EE2B 4EE2C	40	0	電機二甲乙丙合開,預選40人

Course Plan：

Instructor Ku, Chiung-Shen　Office Hours

Email johnsonku@ntut.edu.tw

Last Updated 2024-12-17 00:37:26

Course Objective 本課程包含深度學習簡介，內容講述深度學習概述、應用領域、深度學習與傳統機器學習的區別。 1. TensorFlow介紹：TensorFlow簡介、TensorFlow環境搭建、TensorFlow基礎操作。 2. 神經網路模型：多層感知器（Multilayer Perceptron）、卷積神經網路（Convolutional Neural Network, CNN）、循環神經網路（Recurrent Neural Network, RNN）等。 3. 訓練與評估：模型訓練、超參數調整、評估技巧。 4. 深度學習實戰：使用TensorFlow實現數據分類、影像分類、語音識別等深度學習應用實例。 5. 深度學習進階：深度學習進階技巧，包括半監督學習、生成對抗網路（Generative Adversarial Network, GAN）、強化學習等。 I. Introduction to Deep Learning Definition and history of deep learning Overview of deep learning applications Types of neural networks (Feedforward, Convolutional, Recurrent) II. Neural Network Basics Neurons and activation functions Cost functions and optimization algorithms Overfitting and regularization III. Convolutional Neural Networks (CNNs) Convolution operation Pooling layers Transfer learning with pre-trained models Applications of CNNs (image classification, object detection, segmentation) IV. Recurrent Neural Networks (RNNs) Time-series data and sequential modeling Types of RNNs (Simple RNN, LSTM, GRU) Applications of RNNs (text classification, language translation, speech recognition) V. Generative Adversarial Networks (GANs) Definition and principles of GANs Generator and discriminator networks Applications of GANs (image synthesis, style transfer) VI. TensorFlow Overview of TensorFlow Building and training neural networks in TensorFlow Advanced TensorFlow features (custom layers, GPU acceleration, TensorBoard) VII. Case Studies and Projects Hands-on projects using TensorFlow and deep learning Case studies from various industries (computer vision, natural language processing, recommendation systems) VIII. Conclusion Recap of key concepts and deep learning techniques Discussion of current trends and future directions in deep learning Preparing for a career in deep learning and AI.

Course Schedule Week 1-2: Introduction to Deep Learning Definition and history of deep learning Overview of deep learning applications Types of neural networks (Feedforward, Convolutional, Recurrent) Week 3-4: Neural Network Basics Neurons and activation functions Cost functions and optimization algorithms Overfitting and regularization Week 5-6: Convolutional Neural Networks (CNNs) Convolution operation Pooling layers Transfer learning with pre-trained models Applications of CNNs (image classification, object detection, segmentation) Week 7-8: Recurrent Neural Networks (RNNs) Time-series data and sequential modeling Types of RNNs (Simple RNN, LSTM, GRU) Applications of RNNs (text classification, language translation, speech recognition) Week 9-10: Generative Adversarial Networks (GANs) Definition and principles of GANs Generator and discriminator networks Applications of GANs (image synthesis, style transfer) Week 11-12: TensorFlow Overview of TensorFlow Building and training neural networks in TensorFlow Advanced TensorFlow features (custom layers, GPU acceleration, TensorBoard) Week 13-14: Case Studies and Projects (Part 1) Hands-on projects using TensorFlow and deep learning Case studies from various industries (computer vision, natural language processing, recommendation systems) Week 15-16: Case Studies and Projects (Part 2) Continued hands-on projects using TensorFlow and deep learning Further exploration of case studies and real-world applications Week 17-18: Conclusion Recap of key concepts and deep learning techniques Discussion of current trends and future directions in deep learning Preparing for a career in deep learning and AI.

Evaluation and grading policy 期中作品/Midterm report 30% 期末作品/Final report 30% 平時作業/Homework 40%

Materials All right reserved. Do not reproduce without permission.

Foreign-language textbooks:No
李金洪, 極詳細＋超深入：最新版TensorFlow 1.x/2.x完整工程實作, 深智數位黃士嘉林邑撰 ,輕鬆學會Google TensorFlow 2﹒0人工智慧深度學習實作開發（第二版）博碩文化, 2020

The access to curricular consultation Office hour or making an appoinment by email

The course corresponds to the SDGs SDG1：消除貧窮（No Poverty）
SDG3：良好健康和福祉（Good Health and Well-being）

Note
　

Instructor	Ku, Chiung-Shen　Office Hours
Email	johnsonku@ntut.edu.tw
Last Updated	2024-12-17 00:37:26
Course Objective	本課程包含深度學習簡介，內容講述深度學習概述、應用領域、深度學習與傳統機器學習的區別。 1. TensorFlow介紹：TensorFlow簡介、TensorFlow環境搭建、TensorFlow基礎操作。 2. 神經網路模型：多層感知器（Multilayer Perceptron）、卷積神經網路（Convolutional Neural Network, CNN）、循環神經網路（Recurrent Neural Network, RNN）等。 3. 訓練與評估：模型訓練、超參數調整、評估技巧。 4. 深度學習實戰：使用TensorFlow實現數據分類、影像分類、語音識別等深度學習應用實例。 5. 深度學習進階：深度學習進階技巧，包括半監督學習、生成對抗網路（Generative Adversarial Network, GAN）、強化學習等。 I. Introduction to Deep Learning Definition and history of deep learning Overview of deep learning applications Types of neural networks (Feedforward, Convolutional, Recurrent) II. Neural Network Basics Neurons and activation functions Cost functions and optimization algorithms Overfitting and regularization III. Convolutional Neural Networks (CNNs) Convolution operation Pooling layers Transfer learning with pre-trained models Applications of CNNs (image classification, object detection, segmentation) IV. Recurrent Neural Networks (RNNs) Time-series data and sequential modeling Types of RNNs (Simple RNN, LSTM, GRU) Applications of RNNs (text classification, language translation, speech recognition) V. Generative Adversarial Networks (GANs) Definition and principles of GANs Generator and discriminator networks Applications of GANs (image synthesis, style transfer) VI. TensorFlow Overview of TensorFlow Building and training neural networks in TensorFlow Advanced TensorFlow features (custom layers, GPU acceleration, TensorBoard) VII. Case Studies and Projects Hands-on projects using TensorFlow and deep learning Case studies from various industries (computer vision, natural language processing, recommendation systems) VIII. Conclusion Recap of key concepts and deep learning techniques Discussion of current trends and future directions in deep learning Preparing for a career in deep learning and AI.
Course Schedule	Week 1-2: Introduction to Deep Learning Definition and history of deep learning Overview of deep learning applications Types of neural networks (Feedforward, Convolutional, Recurrent) Week 3-4: Neural Network Basics Neurons and activation functions Cost functions and optimization algorithms Overfitting and regularization Week 5-6: Convolutional Neural Networks (CNNs) Convolution operation Pooling layers Transfer learning with pre-trained models Applications of CNNs (image classification, object detection, segmentation) Week 7-8: Recurrent Neural Networks (RNNs) Time-series data and sequential modeling Types of RNNs (Simple RNN, LSTM, GRU) Applications of RNNs (text classification, language translation, speech recognition) Week 9-10: Generative Adversarial Networks (GANs) Definition and principles of GANs Generator and discriminator networks Applications of GANs (image synthesis, style transfer) Week 11-12: TensorFlow Overview of TensorFlow Building and training neural networks in TensorFlow Advanced TensorFlow features (custom layers, GPU acceleration, TensorBoard) Week 13-14: Case Studies and Projects (Part 1) Hands-on projects using TensorFlow and deep learning Case studies from various industries (computer vision, natural language processing, recommendation systems) Week 15-16: Case Studies and Projects (Part 2) Continued hands-on projects using TensorFlow and deep learning Further exploration of case studies and real-world applications Week 17-18: Conclusion Recap of key concepts and deep learning techniques Discussion of current trends and future directions in deep learning Preparing for a career in deep learning and AI.
Evaluation and grading policy	期中作品/Midterm report 30% 期末作品/Final report 30% 平時作業/Homework 40%
Materials	All right reserved. Do not reproduce without permission. Foreign-language textbooks:No 李金洪, 極詳細＋超深入：最新版TensorFlow 1.x/2.x完整工程實作, 深智數位黃士嘉林邑撰 ,輕鬆學會Google TensorFlow 2﹒0人工智慧深度學習實作開發（第二版）博碩文化, 2020
The access to curricular consultation	Office hour or making an appoinment by email
The course corresponds to the SDGs	SDG1：消除貧窮（No Poverty） SDG3：良好健康和福祉（Good Health and Well-being）
Note