Advanced Machine Learning (FutureLearn)

You will discover where machine learning techniques are used in the data science project workflow. You will then look in detail at supervised learning statistical modeling algorithms for classification and regression problems, examining how these algorithms are related, and how models generated by them can be tuned and evaluated.

You will also look at feature engineering and how to analyse sufficiency of data.

What topics will you cover?

- Statistical Machine Learning Theory

- Analysis and Evaluation of Statistical Models

- Analysis of Data

- Supervised Learning - Artificial Neural Networks

- Supervised Learning - Kernel Methods

- Unsupervised Learning - Clustering

- Unsupervised Learning - Topic Modeling

- Feature Engineering

- Missing Data

- Basic Reinforcement Learning

- Basic Semi-Supervised Learning

What will you achieve?

-By the end of the course, you'll be able to...

- Explain the steps of a typical data science problem, and perform those steps identified as falling under the responsibility of a machine learning specialist.

- Perform a range of pre-processing steps, including feature engineering and management of missing data, as well as explain the utility and importance of such methods.

- Apply a range of advanced machine learning techniques from all major areas of machine learning (supervised, unsupervised, semi-supervised and reinforcement learning) including tuning and regularizing these models.

- Explain how these techniques work, including the relationship between more advanced methods and the simpler methods they are built upon.

- Evaluate rigorously the performance of statistical models, and justify the selection of particular models for use.

- Evaluate rigorously the sufficiency of and suitability of data for a given modelling task