Summary of course topics Supervised learning – labeled data Linear regression Logistic regression Neural networks Support vector machines Unsupervised learning – unlabeled data K-means PCA Anomaly detection Special applications/topics Recommender systems Large scale machine learning Advice on building machine learning systems Bias and variance Regularization What to do next when developing a Read more…
Problem description and pipeline Case study focused around photo OCR Three reasons to do this 1) Look at how a complex system can be put together 2) The idea of a machine learning pipeline What to do next How to do it 3) Some more interesting ideas Applying machine learning to tangible problems Artificial data synthesis What is the photo OCR Read more…
Learning with large datasets This set of notes look at large scale machine learning – how do we deal with big datasets? If you look back at 5-10 year history of machine learning, ML is much better now because we have much more data However, with this increase in data comes Read more…
Recommender systems – introduction Two motivations for talking about recommender systems Important application of ML systems Many technology companies find recommender systems to be absolutely key Think about websites (amazon, Ebay, iTunes genius) Try and recommend new content for you based on passed purchase Substantial part of Amazon’s revenue generation Improvement in Read more…
Anomaly detection – problem motivation Anomaly detection is a reasonably commonly used type of machine learning application Can be thought of as a solution to an unsupervised learning problem But, has aspects of supervised learning What is anomaly detection? Imagine you’re an aircraft engine manufacturer As engines roll off your assembly line you’re doing QA Read more…
Motivation 1: Data compression Start talking about a second type of unsupervised learning problem – dimensionality reduction Why should we look at dimensionality reduction? Compression Speeds up algorithms Reduces space used by data for them What is dimensionality reduction? So you’ve collected many features – maybe more than you need Can you “simply” Read more…
Unsupervised learning – introduction Talk about clustering Learning from unlabeled data Unsupervised learning Useful to contras with supervised learning Compare and contrast Supervised learning Given a set of labels, fit a hypothesis to it Unsupervised learning Try and determining structure in the data Clustering algorithm groups data together based on data features What Read more…
Support Vector Machine (SVM) – Optimization objective So far, we’ve seen a range of different algorithms With supervised learning algorithms – performance is pretty similar What matters more often is; The amount of training data Skill of applying algorithms One final supervised learning algorithm that is widely used – support vector machine (SVM) Compared Read more…
Machine learning systems design In this section we’ll touch on how to put together a system Previous sections have looked at a wide range of different issues in significant focus This section is less mathematical, but material will be very useful non-the-less Consider the system approach You can understand all Read more…
Deciding what to try next We now know many techniques But, there is a big difference between someone who knows an algorithm vs. someone less familiar and doesn’t understand how to apply them Make sure you know how to chose the best avenues to explore the various techniques Here we focus deciding Read more…