CS 60050 Machine Learning. What is Machine Learning?. Adapt to / learn from data To optimize a performance function Can be used to: Extract knowledge from data Learn tasks that are difficult to formalise Create software that improves over time. When to learn
Can be used to:
Applications are diverse but methods are generic
Eg. Classification, regression, prediction, function approx.
Eg. Clustering, visualisation, dimensionality reduction
Eg. Speech recognition, DNA data analysis
Machine learning methods can be unified within the framework of statistical learning:
Learning machines need to do both.
Inductive learning hypothesis:
“If the hypothesis works for enough data
then it will work on new examples.”
Data representation: Greyscale images
Task: Classification (0,1,2,3…..9)
Data representation: features from spectral analysis of speech signals (two in this simple example).
Task: Classification of vowel sounds in words of the form “h-?-d”
Data representation: ~10000 Green/red intensity levels ranging from 10-10000.
Tasks: Sample classification, gene classification, visualisation and clustering of genes/samples.
effectively separates cancer subtypes.
A large part of the module will deal with methods
that have an explicit probabilistic interpretation:
eg. is a handwritten digit a three or an eight ?
E. Alpaydin’s “Introduction to Machine Learning”
T. Mitchell’s “Machine Learning”