Building Deep Learning Applications

  NTUT Deep Learning FB Group Playlist

Week Topic Learning Objectives Slides Code Video
Course Requirements
  • Kaggle homework (40%)
  • Midterm & Final exam (20% & 15%)
  • Final project (20%)
  • Quizzes (5%)
  • Bonus (5%)
pdf
Text Book François Chollet, Deep Learning with Python, 2nd Edition, Manning, 2021 GitHub
Reference Ian Goodfellow and Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016 slides
0 Past, Present, and Future of AI
  • Free your imagination to unleash your potential!
1 Introduction to Deep Learning
  • What is the Machine Learning?
  • Neural Networks, Gradient Descent and Backpropagation
  • State-of-the-arts of deep learning
pdf
2 Applied Math Linear Algebra
  • Scalar, vector, matrix, tensor
  • Matrix multiplication
  • Dot product and outer product
  • Lp norm
pdf
Probability
  • Mutually exclusive & independence
  • Conditional Probability & Bayes Theorem
  • Naive Bayes classifier
pdf
3 Introduction to Keras
  • Write Keras code on Google Colab
  • Create a simple Dense Neural Networks
  • Use DNN to solve classification and regression problems
  • Batch, Epoch and Learning rate
pdf IMDB_review
financial_news
house_pricing
4 Machine Learning Basics
  • Five tribes of Machine Learning
  • KNN, Perceptron, SVM
  • Logistic Regression
  • Linear, Ridge, Lasso Regression and ElasticNet
  • Regularization
  • Overfitting, Cross-validation, Dropout
  • Confusion Matrix, Accuracy, Precision, Recall
pdf classifiers regressors
5 Convolutional Neural Networks (CNNs)
  • Convolution Filters
  • Kernel vs. Filter
  • Using pre-trained models
  • AlexNet, Inception, ResNet, Xception
  • NAS & EfficientNet
pdf
6 Advanced Keras API
  • Functional API
  • Multi-input and Multi-output model
  • Depthwise Separable Convolution
pdf
7 NLP & Word Embedding
  • Natural Language Processing
  • N-gram & Skip-gram Model
  • Word2vec
  • GloVe
pdf
8 RNN & LSTM
  • Recurrent Neural Networks
  • LSTM
  • Temporal CNN
pdf
9 Attention & Transformer
  • Attention is All You Need
  • Self-attention
  • Transformer
pdf
Modern NLP Models
  • Bidirectional Encoder Representations from Transformers (BERT)
  • Generative Pre-trained Transformers (GPT)
  • Embeddings from Language Models (ELMo)
pdf
10 Generative Deep Learning
  • Generating Text using LSTM
  • Neural Style Transfer
  • Variational Auto Encoder
pdf
Generative Adversarial Networks (GANs)
  • Generative Adversarial Networks
  • Conditional GAN
  • CycleGAN
  • GauGAN, Adversarial Attack
pdf
11 Object Detection and Segmentation
  • Region Proposal & RCNN
  • YOLO v1 - v5
  • SSD, FPN, Retina Net, EfficientDet
  • U-Net, Mask RCNN
pdf Run YOLOv4
12 Deep Reinforcement Learning
  • What is Reinforcement Learning?
  • State, Action, Reward and Policy
  • Value-based vs. Policy-based Learning
  • On-policy vs. Off-policy
pdf
13 Action Recognition
  • Optical Flow
  • Two-stream Model
  • C3D, TSN, and other SOTA Models
pdf
14 Deep Learning on Graphs
  • Graphs and Networks
  • Vertices, Edges, Degree
  • Deep Random Walk
  • Graph Convolutions
pdf
Final Project Demo