MLSpecialization_DeepLearningIA

Contains Solutions and Notes for the Machine Learning Specialization by Andrew NG on Coursera

Course 1 : Supervised Machine Learning: Regression and Classification

• Week 1

  • Regression
  • Supervised vs unsupervised learning
  • Model Representation
  • Cost Function
  • Gradient Descent

• Week 2

  • Gradient descent.
  • Multiple linear regression
  • Numpy Vectorization
  • Multi Variate Regression
  • Feature Scaling
  • Feature Engineering
  • Linear Regression

• Week 3

  • Cost function and gradient descent for logistic regression
  • Classification
  • Sigmoid Function
  • Decision Boundary
  • Logistic Loss
  • Overfitting
  • Regularization
  • Logistic Regression

Certificate Of Completion

Course 2 : Advanced Learning Algorithms

• Week 1
  • Neural networks model and intuition
  • TensorFlow implementation
  • Neural Networks Implementation in Numpy
  • Neurons and Layers
  • Neural Networks for Binary Classification

• Week 2
  • Neural Networks Training
  • Activation Functions
  • Multiclass Classification
  • RElu
  • Softmax
  • Neural Networks For Handwritten Digit Recognition

• Week 3
  • Bias and Variance
  • Machine Learning Development Process

• Week 4
  • Decision Trees
  • Decision Trees Learning
  • Decision Trees Ensembles

Certificate of Completion

Course 3 : Unsupervised Learning, Recommenders, Reinforcement Learning

• Week 1
  • Clustering
  • Anomaly Detection
  • K means
  • Anomaly Detection

• Week 2
  • Collaborative Filtering
  • Recommender systems implementation
  • Content-based filtering
  • Collaborative Filtering RecSys
  • RecSys using Neural Networks

• Week 3
  • Reinforcement learning
  • State-action value function
  • Continuous state spaces
  • Deep Q-Learning - Lunar Lander Example

Certificate of Completion

Specialization Certificate

Course Review :

This Course is a great place to start and get into Machine Learning algorithms.

Special thanks to Professor Andrew Ng for structuring and tailoring this Course.

---

Some results :

• Write an unsupervised learning algorithm to Land the Lunar Lander Using Deep Q-Learning

  • The Rover was trained to land correctly on the surface, correctly between the flags as indicators after many unsuccessful attempts in learning how to do it.
  • The final landing after training the agent using appropriate parameters :

lunar_lander.mp4

• Write an algorithm for a Movie Recommender System

  • A movie database is collected based on its genre.
  • A content based filtering and collaborative filtering algorithm is trained and the movie recommender system is implemented.
  • It gives movie recommendentations based on the movie genre.

Bugs and feature requests

Have a bug or a feature request? Please first read the issue guidelines and search for existing and closed issues. If your problem or idea is not addressed yet, please open a new issue.

Creators

• https://github.com/mvanzulli

Copyright and license

Code and documentation copyright 2022 the authors. Code released under the MIT License.