Machine Learning Specialization

What the course covers

Summarised by TrustMyCourse from public course information. Wording is our own, not the provider’s.

1. 01

   Introduction to machine learning

   Course 1, week 1. Supervised and unsupervised learning, linear regression, cost functions and gradient descent.

2. 02

   Regression with multiple input variables

   Course 1, week 2. Vectorisation, feature scaling, feature engineering and polynomial regression.

3. 03

   Classification

   Course 1, week 3. Logistic regression, decision boundaries, and overfitting addressed through regularisation.

4. 04

   Neural networks

   Course 2, week 1. Building networks in TensorFlow, then implementing the same thing from scratch in Python.

5. 05

   Neural network training

   Course 2, week 2. Activation functions beyond sigmoid, multiclass classification with softmax, and the Adam optimiser.

6. 06

   Advice for applying machine learning

   Course 2, week 3. Diagnosing bias and variance, learning curves, error analysis, transfer learning, and fairness considerations.

7. 07

   Decision trees

   Course 2, week 4. Tree fundamentals, information gain, random forests and XGBoost, and choosing between algorithms.

8. 08

   Unsupervised learning

   Course 3, week 1. K-means clustering and anomaly detection using Gaussian distributions, around 9 hours.

9. 09

   Recommender systems

   Course 3, week 2. Collaborative filtering, content-based deep learning methods, PCA, and the ethics of recommendation, around 11 hours.

10. 10

    Reinforcement learning

    Course 3, week 3. Reinforcement learning fundamentals and building a deep Q-learning network, around 8 hours.