1. Introduction to Machine Learning Fall 2013 Comp3710 Artificial Intelligence Computing Science Thompson Rivers University

2. TRU-COMP3710 Intro to Machine Learning2 Course Outline Part I – Introduction to Artificial Intelligence Part II – Classical Artificial Intelligence Part III – Machine Learning Introduction to Machine Learning Neural Networks Probabilistic Reasoning and Bayesian Belief Networks Artificial Life: Learning through Emergent Behavior Part IV – Advanced Topics

3. TRU-COMP3710 Intro to Machine Learning3 Learning Objectives Define what classification is. List the three types of attributes with examples. Summarize what concept learning is. Compute the information gain for an attribute from a given training data set. Construct a decision tree from a given training data set, using information gains....

4. TRU-COMP3710 Intro to Machine Learning4 Chapter Outline 1. Introduction Introduction 2. Training Training 3. Rote Learning Rote Learning 4. Learning Concepts Learning Concepts 5. Inductive Bias Inductive Bias 6. Decision-Tree Induction Decision-Tree Induction 7. The Problem of Overfitting The Problem of Overfitting 8. The Nearest Neighbor Algorithm The Nearest Neighbor Algorithm 9. Learning Neural Networks Learning Neural Networks 10. Supervised Learning Supervised Learning 11. Reinforcement Learning Reinforcement Learning 12. Unsupervised Learning Unsupervised Learning

5. TRU-COMP3710 Intro to Machine Learning5 1. Introduction [Q] What is learning: Learning and intelligence are intimately related to each other. It is usually agreed that a system capable of learning deserves to be called intelligent. [Q] Is it possible to make a machine learn? We will discuss about the concept of learning methods. [Q] What are we going to make a machine learn? Topics

6. TRU-COMP3710 Intro to Machine Learning6 2. Training [Q] How do we learn? Learning problems usually involve classifying inputs into a set of classifications. Example of the people in this class [Q] Can we classify us? Learning is only possible if there is a relationship between the data and the classifications. [Q] What kind of relationship? How to classify? Based on similarity Training involves providing the system with data which has been manually classified. Learning systems use the training data to learn to classify unseen data.

7. TRU-COMP3710 Intro to Machine Learning7 3 data types based on comparison methods The input data usually consist of multiple attributes. The attributes are not all of the same type. Numeric or numerical: E.g., 0, 1,.67,... ; order and distance between two values Ordinal: E.g., Small, Medium, Large, Extra Large; only order Categorical (Boolean can be considered this type.): English, Chines, Spanish; not even order Example of the people in this class Agenumerical, ordinal, or categorical? Height? Weight? Mother tongue? Hair color? Number of legs? Number of armsIs it really necessary to classify us? Number of nose... Topics

8. TRU-COMP3710 Intro to Machine Learning8 3. Rote Learning Simply involves memorizing the classifications of the training data. A very simple learning method. Can only classify previously seen data – unseen data cannot be classified by a rote learner. [Q] Is this learning type good enough? What if we have a similar data? Topics

9. TRU-COMP3710 Intro to Machine Learning9 4. Concept Learning Concept learning involves determining a mapping from a set of input variables to a Boolean value. Such methods are known as inductive learning methods. If a function can be found which maps training data to correct classifications, then it will also work well for unseen data – hopefully! This process is known as generalization. A simple toy problem to determine whether driving in a particular manner in particular road conditions is safe or not: AttributePossible values SpeedSlow, medium, fastType? WeatherWind, rain, snow, sunType? Distance from car in front10ft, 20ft, 20ft,...Type? Units of alcohol driver has drunk0, 1, 2, 3, 4, 5Type? Time of dayMorning, afternoon, evening, night TemperatureCold, worm, hot

10. TRU-COMP3710 Intro to Machine Learning10 A hypothesis (or object) is a vector (or list) of attributes: h 1 = h 2 = ? means “we do not care”, i.e., any value This looks clearly untrue.-> Negative training example h 3 =  means no value In concept learning, a training hypothesis is either a positive or negative (true or false) (or multiple classes). Concept learning can be thought as search through a search space that consists of hypotheses, where the goal is the hypothesis that is most closely mapped to a given query. [Q] How to define “closely”? Topics

11. TRU-COMP3710 Intro to Machine Learning11 5. Inductive Bias All learning methods have an inductive bias. Inductive bias refers to the restrictions that are imposed by the assumptions made in the learning method. E.g., the solution to the problem of road safety can be expressed as a conjunction of a set of six concepts. This does not allow for more complex expressions that cannot be expressed as a conjunction. Therefore there can be some potential solutions that we cannot explore. However, without inductive bias, a learning method could not learn to generalize. Occam’s razor is an example of an inductive bias: The best hypothesis to select is the simplest one. h 1 = h 2 = Which one is better? Topics

12. Box-office success problem Training data set: [Q] How to obtain this kind of table? Query: (Thailand, Yes, Romance) -> Success? [Q] How many comparisons when brute-force search is used? Let’s try. TRU-COMP3710 Intro to Machine Learning12 6. Decision Tree Induction FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue

13. TRU-COMP3710 Intro to Machine Learning13 In the box-office success problem, what must be the first question? Country, Big Start, or Genre? Country is a significant determinant of whether a film will be a success or not. Hence the first question is Country. What is the next then? Using a tree? [Q] How to determine which factor is the most significant determinant? FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue

14. A decision tree takes an input and gives a Boolean output (or a class). Box-office success problem:  Decision trees can represent more complex expressions, involving disjunctions and conjunctions.  ((Country = USA)  (Big Star = Yes))  ((Country = Europe)  (Genre = Comedy))  All the objects in the training data set have classes. In this example, true and false.  [Q] IF-THEN rules?  [Q] Can we use decision tree induction as an expert system? TRU-COMP3710 Intro to Machine Learning14

15. TRU-COMP3710 Intro to Machine Learning15 Decision tree induction involves creating a decision tree from a set of training data that can be used to correctly classify the training data. A query will be given without class to find the class to which the query fits most. [Q] How to create an efficient decision tree from a training data set? ID3, developed in the 1980s, is an example of a decision tree learning algorithm. ID3 builds the decision tree from the top down, selecting the attributes (or also called features) from the training data that provide the most information at each stage. That is, most determinant attribute.

16. TRU-COMP3710 Intro to Machine Learning16 [Q] In the box-office success problem, what must be the first question? Country, Big Start, or Genre? Country is a significant determinant of whether a film will be a success or not. Hence the first question is Country. [Q] How do we know what the most significant determinant is? FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue

17. TRU-COMP3710 Intro to Machine Learning17 l ID3 selects attributes based on information gain. l Information gain is the reduction in entropy caused by a decision. l In information theory, entropy is a measure of the uncertainty associated with a random variable. l Entropy of a set is defined as: H(S) = - p 0 log 2 p 0 - p 1 log 2 p 1 -... (if there are other classes) o p 0 is the proportion of the training data that are positive (class 0) examples. o p 1 is the proportion which are negative (class 1) examples. l [Q] The most certain case? o The entropy of S is 0 when all the examples are positive, or when all the examples are negative. l [Q] The most ambiguous case? o The entropy reaches its maximum value of 1 when exactly half of the examples are positive, and half are negative. l [Q] The smaller, the better?

18. TRU-COMP3710 Intro to Machine Learning18 The information gain of a particular attribute tells us how closely the attribute represents the entire target function, and so at each stage, the attribute that gives the highest information gain is chosen to turn into a question. [Q] How to obtain the information gain of a particular attribute? Gain = 1 –  (the weighted entropy for each value in the attribute) Then, 0 <= Gain <= 1

19. TRU-COMP3710 Intro to Machine Learning19 [Q] The information gain for Country? Gain = 1 – w-entropy(USA) – w-entropy(Europe) – w-entropy(Other) For USA, there are 4 films. 3 out of 4 are True. H(USA) = – 3/4 log 2 (3/4) – (1/4) log 2 (1/4) =.811 Similarly, (2/4 are True, and 2/4 are False for Europe; 2/2 are False for Other) H(Europe) = 1; H(Other) = 0 The weights for USA, Europe and Other are 4/10, 4/10 and 2/10. Gain = 1 – (4/10 *.811) – (4/10 * 1) – (2/10 * 0) =.2756 FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue H(S) = - p 0 log 2 p 0 - p 1 log 2 p 1

20. TRU-COMP3710 Intro to Machine Learning20 [Q] The information gain for Big Star? For Yes, there are 7 films. 4 out of 7 are True. H(Yes) = –4/7 log 2 (4/7) – (3/7) log 2 (3/7) = ? Similarly, H(No) = ? The weights for Yes and No are 7/10 and 3/10. Gain = 1 – (7/10 * ?) – (3/10 * ?) =.01 Gain for Genre =.17 FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue H(S) = - p 0 log 2 p 0 - p 1 log 2 p 1

21. TRU-COMP3710 Intro to Machine Learning21 The information gain for Country=.2756 The information gain for Big Star =.01 The information gain for Genre =.17 Therefore, the attribute Country provides the greatest information gain and so is placed at the top of the decision tree. This method is then applied recursively to the sub-branches of the tree. For example of USA, we need to decide the most determinant among BigStar and Genre. FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 10USAYesComedyTrue

22. TRU-COMP3710 Intro to Machine Learning 22 The case of Country = Other after Country is selected: The information gain for Big Star: For Yes, there is 1 film. 1 out 1 is False. -> H(Yes) = 0 For No, there is 1 film. 1 out 1 is False. -> H(No) = 0 Gain = 1 – 1/2 * 0 – 1/2 * 0 = 1 Similarly, the information gain for Genre = 1 Hence there is no more branch in this case. [Q] The meaning is ??? This method is then applied recursively to other sub-branches of the tree. FilmCountryBig StarGenreSuccess Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Topics [Q] No need to check Genre?

23. TRU-COMP3710 Intro to Machine Learning23 7. The Problem of Overfitting Noise can be in the training data set The training data do not adequately represent the entire space of possible data. Then, for example, a decision tree can perform poorly at classifying unseen data. Not only decision trees, but also other learning methods

24. TRU-COMP3710 Intro to Machine Learning24 Black dots represent positive examples, white dots negative. The two lines represent two different hypotheses. In the first diagram, there are just a few items of training data, correctly classified by the hypothesis represented by the darker line. In the second and third diagrams we see the complete set of data, and that the simpler hypothesis which matched the training data less well matches the rest of the data better than the more complex hypothesis, which overfits. noise Topics

25. TRU-COMP3710 Intro to Machine Learning25 8. The Nearest Neighbor Algorithm [Q] Any other learning method? The Nearest Neighbor algorithm is an example of instance based learning. Instance based learning involves storing training data and using it to attempt to classify new data as it arrives. FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 2USANoComedyFalse Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse Film 9EuropeYesComedyTrue Film 10USAYesComedyTrue FilmCountryBig StarGenreSuccess Film 1USAYesSFTrue Film 3USAYesComedyTrue Film 4EuropeNoComedyTrue Film 9EuropeYesComedyTrue Film 2USANoComedyFalse Film 5EuropeYesSFFalse Film 5EuropeYesSFFalse Film 6EuropeYesRomanceFalse Film 7OtherYesComedyFalse Film 8OtherNoSFFalse QueryThailandYesRomance???

26. TRU-COMP3710 Intro to Machine Learning26 The nearest neighbor algorithm works with data that consists of vectors of numerical attributes. Each vector represents a point in n-dimensional space. When an unseen data item is to be classified, the Euclidean distance is calculated between this item and all training data. For example, the distance between and is: [Q] How to classify?

27. TRU-COMP3710 Intro to Machine Learning27 [Q] How to classify, when k = 3? The k-nearest neighbor algorithm: The classification for the unseen data is usually selected as the one that is most common amongst the k-nearest neighbors. Shepard’s method: This involves allowing all training data to contribute to the classification with their contribution being proportional to their distance from the data item to be classified. For each class,  1/d i... Advantage: Unlike decision tree learning, the nearest neighbor algorithm performs very well with noisy input data. Disadvantage: [Q] But, what if the training data set is huge? Any good idea? Topics

28. TRU-COMP3710 Intro to Machine Learning28 9. Neural Networks An artificial neural network is a network of artificial neurons, which is based on the operation of the human brain. Neural networks usually have their nodes arranged in layers. One layer is the input layer, and another is an output layer. There are one or more hidden layers between these two.

29. TRU-COMP3710 Intro to Machine Learning29 The connections between nodes have weights associated with them, which determine the behavior of the network. Input data is applied to the input layer. Neurons fire if their inputs are above a certain level. If one neuron is connected to another the firing of one may cause the firing of the next. In the next unit, we will discuss them in more detail. Topics

30. TRU-COMP3710 Intro to Machine Learning30 10. Supervised Learning Two types of learning Supervised, Unsupervised Supervised learning learns by being presented with pre-classified training data set. [Q] Is decision tree induction supervised learning? [Q] Is the k-nearest neighbor algorithm supervised learning? Many neural networks use supervised learning. Pre-classified training data is provided to the network before it is presented with unseen data. The training data causes the weights in the network to be set to levels such that unseen data can be classified correctly. Neural networks are able to learn to classify extremely complex functions. Demo: http://www.cbu.edu/~pong/ai/hopfield/hopfieldapplet.htmlhttp://www.cbu.edu/~pong/ai/hopfield/hopfieldapplet.html [Q] Can the positive/negative feedback information be used in training? Topics

31. TRU-COMP3710 Intro to Machine Learning31 11. Reinforcement Learning Systems that learn using reinforcement learning are given a positive feedback when they classify data correctly, and negative feedback when they classify data incorrectly. Credit assignment is needed to reward nodes in a network correctly. Topics

32. TRU-COMP3710 Intro to Machine Learning32 12. Unsupervised Learning Unsupervised learning learns without any training data set. Unsupervised learning networks learn without requiring human intervention. No training data is required. [Q] How is that possible? The system learns to cluster input data into a set of classifications that are not previously defined. This is called clustering. Clustering is a basic tool for data mining and pattern recognition. Example: Fuzzy C-Means, EM, Kohonen Maps.... Topics