1. CS 60050 Machine Learning

2. 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

3. When to learn Human expertise does not exist (navigating on Mars) Humans are unable to explain their expertise (speech recognition) Solution changes in time (routing on a computer network) Solution needs to be adapted to particular cases (user biometrics) Learning involves Learning general models from data Data is cheap and abundant. Knowledge is expensive and scarce Customer transactions to computer behaviour Build a model that is a good and useful approximation to the data

4. Applications Speech and hand-writing recognition Autonomous robot control Data mining and bioinformatics: motifs, alignment, … Playing games Fault detection Clinical diagnosis Spam email detection Credit scoring, fraud detection Web mining: search engines Market basket analysis, Applications are diverse but methods are generic

5. Generic methods Learning from labelled data (supervised learning) Eg. Classification, regression, prediction, function approx. Learning from unlabelled data (unsupervised learning) Eg. Clustering, visualisation, dimensionality reduction Learning from sequential data Eg. Speech recognition, DNA data analysis Associations Reinforcement Learning

6. Statistical Learning Machine learning methods can be unified within the framework of statistical learning:  Data is considered to be a sample from a probability distribution.  Typically, we don’t expect perfect learning but only “probably correct” learning.  Statistical concepts are the key to measuring our expected performance on novel problem instances.

7. Induction and inference Induction: Generalizing from specific examples. Inference: Drawing conclusions from possibly incomplete knowledge. Learning machines need to do both.

8. Inductive learning Data produced by “target”. Hypothesis learned from data in order to “explain”, “predict”,“model” or “control” target. Generalisation ability is essential. Inductive learning hypothesis: “If the hypothesis works for enough data then it will work on new examples.”

9. Example 1: Hand-written digits Data representation: Greyscale images Task: Classification (0,1,2,3…..9) Problem features: Highly variable inputs from same class including some “weird” inputs, imperfect human classification, high cost associated with errors so “don’t know” may be useful.

11. Example 2: Speech recognition 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” Problem features: Highly variable data with same classification. Good feature selection is very important. Speech recognition is often broken into a number of smaller tasks like this.

13. Example 3: DNA microarrays DNA from ~10000 genes attached to a glass slide (the microarray). Green and red labels attached to mRNA from two different samples. mRNA is hybridized (stuck) to the DNA on the chip and green/red ratio is used to measure relative abundance of gene products.

15. DNA microarrays Data representation: ~10000 Green/red intensity levels ranging from 10-10000. Tasks: Sample classification, gene classification, visualisation and clustering of genes/samples. Problem features: High-dimensional data but relatively small number of examples. Extremely noisy data (noise ~ signal). Lack of good domain knowledge.

16. Projection of 10000 dimensional data onto 2D using PCA effectively separates cancer subtypes.

17. Probabilistic models A large part of the module will deal with methods that have an explicit probabilistic interpretation: Good for dealing with uncertainty eg. is a handwritten digit a three or an eight ? Provides interpretable results Unifies methods from different fields

18. Relevant disciplines Algorithms Artificial intelligence Control Statistics Information theory Dynamical systems Neurobiology Signal processing Linear algebra Etc, etc ….. Researchers in machine learning come from a variety of backgrounds.

19. Text books E. Alpaydin’s “Introduction to Machine Learning” T. Mitchell’s “Machine Learning”

20. Supervised Learning: Uses Prediction of future cases Knowledge extraction Compression Outlier detection

21. Unsupervised Learning Clustering: grouping similar instances Example applications  Customer segmentation in CRM  Learning motifs in bioinformatics  Clustering items based on similarity  Clustering users based on interests

22. Reinforcement Learning Learning a policy: A sequence of outputs No supervised output but delayed reward Credit assignment problem Game playing Robot in a maze Multiple agnts, partial observability