Advanced Machine Learning Coursera MOOC Specialization

National Research University Higher School of Economics - Yandex

Coursera Webpage

Syllabus

This specialization gives an introduction to deep learning, reinforcement learning, natural language understanding, computer vision and Bayesian methods. Top Kaggle machine learning practitioners and CERN scientists will share their experience of solving real-world problems and help you to fill the gaps between theory and practice. Upon completion of 7 courses you will be able to apply modern machine learning methods in enterprise and understand the caveats of real-world data and settings.

You will master your skills by solving a wide variety of real-world problems like image captioning and automatic game playing throughout the course projects. You will gain the hands-on experience of applying advanced machine learning techniques that provide the foundation to the current state-of-the art in AI.

Table of Contents

1 - Introduction to Deep Learning (certified completion)

• week 1: Optimization
• week 2: Multilayer Perceptron and introduction to Tensorflow/Keras
• week 3: Convolutional Neural Networks
• week 4: Autoencoders and Generative Adversarial Networks
• week 5: Recurrent Neural Networks
• Final Project: Image Captioning

2 - How to Win a Data Science Competition: Learn From Top Kagglers (certified completion)

• week 1: Feature Preprocessing and Engineering
• week 2: Exploratory Data Analysis, Validation Strategies and Data Leakages
• week 3: Metric Optimization and Advanced Feature Engineering I
• week 4: Hyperparameter Optimization, Advanced Feature Engineering II and Ensembling
• Final Project: Kaggle Competition (Predict Future Sales)

3 - Bayesian Methods for Machine Learning (certified completion)

• week 1: Refresher on Bayesian probability theory
• week 2: Expectation-Maximization algorithm and Gaussian Mixture Models
• week 3: Variational Inference and Latent Dirichlet Allocation
• week 4: Markov Chain Monte Carlo
• week 5: Bayesian Neural Networks and Variational Autoencoders
• week 6: Gaussian Processes and Bayesian Optimization
• Final Project: Forensics to generate images of suspects

4 - Natural Language Processing (ON HOLD)

• week 1: Text Classification with Linear Models
• week 2: Language Modelling with Probabilistic Graphical Models and Neural Networks
• week 3: Word Embeddings and Topic Models
• week 4: Machine Translation and Sequence-To-Sequence Models
• Final Project: StackOverflow Task-Oriented Chatbot

5 - Practical Reinforcement Learning (certified completion)

• week 1: Introduction to Reinforcement Learning
• week 2: Model-Based Reinforcement Learning (Dynamic Programming)
• week 3: Model-Free Reinforcement Learning (SARSA, Monte Carlo, Q-Learning)
• week 4: Approximate and Deep Reinforcement Learning (Deep Q-Learning)
• week 5: Policy Gradient Reinforcement Learning
• week 6: Advanced Topics on Exploration and Planning

Future courses

6 - Addressing Large Hadron Collider Challenges by Machine Learning (ON HOLD)

7 - Deep Learning in Computer Vision (ON HOLD)