Machine Learning Specialization

(Course)(Lecture Notes)

Concept Notes

• Course 1 Week 1
  • machine learning
    • supervised learning
    • unsupervised learning
  • machine learning development process
  • learning algorithm
  • learning curves
  • regression - simple regression
  • cost function & loss function
  • gradient descent
• Course 1 Week 2
  • mean squared error
  • multiple regression
• Course 1 Week 3
  • binomial logistic regression
  • regularization
• Course 2 Week 1
  • artificial neural network
    • neuron
    • activation function
  • convolutional neural network
• Course 2 Week 2
  • activation function
    • linear activation function
    • ReLU
  • softmax regression
    • softmax
• Course 2 Week 3
  • Polynomial Regression
  • data augmentation
    • data synthesis
  • error analysis
    • Bias-Variance
      • overfitting
      • underfitting
    • classification metric
      • precision
      • sensitivity (aka recall)
      • F-beta
      • ROC curve
      • area under ROC curve
  • transfer learning
    • supervised pre-training
    • fine tuning
• Course 2 Week 4
  • decision tree (ML)
  • ensemble method
    • bagging
      • Random Forest
    • boosting
      • Adaboost
      • gradient boosting
• Course 3 Week 1
  • unsupervised learning
    • clustering
      • k-means
    • anomaly detection
• Course 3 Week 2
  • recommender system
    • content-based filtering
    • collaborative filtering
      • neighborhood-based collaborative filtering
• Course 3 Week 3
  • reinforcement learning
  • mini-batch gradient descent

Assignments

Course 1

1. Week 2: Linear regression
2. Week 3: Logistic regression

Course 2

1. Week 1: artificial neural network for Binary Classification with sigmoid function
2. Week 2: artificial neural network for Multiclass Classification with softmax regression
3. Week 3: evaluating Polynomial Regression with machine learning development process
4. Week 4: decision tree from scratch

Course 3

1. Week 1: Image compression with k-means
2. Week 2: with neighborhood-based collaborative filtering
3. Week 2: with content-based filtering
4. Week 3: with reinforcement learning

Supplements

Course 1: Regression & Classification

1. StatQuest Playlist: Machine Learning

Course 2: Advanced Learning Algorithms

1. 3Blue1Brown Series: Neural Networks
2. StatQuest Playlist: Neural Networks / Deep Learning
3. Neural Networks and Deep Learning (Nielsen, 2015)
4. Deep Learning Textbook (Goodfellow et al., 2016)