Machine learning

Machine learning · 24. February 2019

Pros and cons of classical supervised ML algorithms

Here we explore the pros and cons of some the most popular classical machine learning algorithms for supervised learning. To recap, this is a learning situation where we are given some labelled data and the model must predict the value or class of a new datapoint using a hypothesis function that it has learned from studying the provided examples. By ‘classical’ machine leaning algorithms I mean anything that is not a neural network. We will cover the advantages and disadvantages of various...

Overfitting, bias-variance and learning curves

We explore what it means for a machine learning model to generalize well. These concepts are important to keep in mind when thinking about all sorts of supervised machine learning problems.

Introduction to classification metrics

Where we take a look at some of the common metrics for evaluating the success of classification models