1. 计算机学院 计算感知 Support Vector Machines

2. 2 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Perceptron Revisited: Linear Separators Binary classification can be viewed as the task of separating classes in feature space: w T x + b = 0 w T x + b < 0 w T x + b > 0 f(x) = sign(w T x + b)

3. 3 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear Separators Which of the linear separators is optimal?

4. 4 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Classification Margin Distance from example x i to the separator is Examples closest to the hyperplane are support vectors. Margin ρ of the separator is the distance between support vectors. r ρ

5. 5 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Maximum Margin Classification Maximizing the margin is good according to intuition and PAC theory. Implies that only support vectors matter; other training examples are ignorable.

6. 6 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear SVM Mathematically Let training set {(x i, y i )} i=1..n, x i  R d, y i  {-1, 1} be separated by a hyperplane with margin ρ. Then for each training example (x i, y i ): For every support vector x s the above inequality is an equality. After rescaling w and b by ρ/2 in the equality, we obtain that distance between each x s and the hyperplane is Then the margin can be expressed through (rescaled) w and b as: w T x i + b ≤ - ρ/2 if y i = -1 w T x i + b ≥ ρ/2 if y i = 1 y i (w T x i + b) ≥ ρ/2 

7. 7 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear SVMs Mathematically (cont.) Then we can formulate the quadratic optimization problem: Which can be reformulated as: Find w and b such that is maximized and for all (x i, y i ), i=1..n : y i (w T x i + b) ≥ 1 Find w and b such that Φ(w) = ||w|| 2 =w T w is minimized and for all (x i, y i ), i=1..n : y i (w T x i + b) ≥ 1

8. 8 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Solving the Optimization Problem Need to optimize a quadratic function subject to linear constraints. Quadratic optimization problems are a well-known class of mathematical programming problems for which several (non-trivial) algorithms exist. The solution involves constructing a dual problem where a Lagrange multiplier α i is associated with every inequality constraint in the primal (original) problem: Find w and b such that Φ(w) =w T w is minimized and for all (x i, y i ), i=1..n : y i (w T x i + b) ≥ 1 Find α 1 …α n such that Q( α ) = Σ α i - ½ ΣΣ α i α j y i y j x i T x j is maximized and (1) Σ α i y i = 0 (2) α i ≥ 0 for all α i

9. 9 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 The Optimization Problem Solution Given a solution α 1 …α n to the dual problem, solution to the primal is: Each non-zero α i indicates that corresponding x i is a support vector. Then the classifying function is (note that we don’t need w explicitly): Notice that it relies on an inner product between the test point x and the support vectors x i – we will return to this later. Also keep in mind that solving the optimization problem involved computing the inner products x i T x j between all training points. w = Σ α i y i x i b = y k - Σ α i y i x i T x k for any α k > 0 f(x) = Σ α i y i x i T x + b

10. 10 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Soft Margin Classification What if the training set is not linearly separable? Slack variables ξ i can be added to allow misclassification of difficult or noisy examples, resulting margin called soft. ξiξi ξiξi

11. 11 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Soft Margin Classification Mathematically The old formulation: Modified formulation incorporates slack variables: Parameter C can be viewed as a way to control overfitting: it “trades off” the relative importance of maximizing the margin and fitting the training data. Find w and b such that Φ(w) =w T w is minimized and for all (x i,y i ), i=1..n : y i (w T x i + b) ≥ 1 Find w and b such that Φ(w) =w T w + C Σ ξ i is minimized and for all (x i,y i ), i=1..n : y i (w T x i + b) ≥ 1 – ξ i,, ξ i ≥ 0

12. 12 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Soft Margin Classification – Solution Dual problem is identical to separable case (would not be identical if the 2- norm penalty for slack variables CΣξ i 2 was used in primal objective, we would need additional Lagrange multipliers for slack variables): Again, x i with non-zero α i will be support vectors. Solution to the dual problem is: Find α 1 …α N such that Q( α ) = Σ α i - ½ ΣΣ α i α j y i y j x i T x j is maximized and (1) Σ α i y i = 0 (2) 0 ≤ α i ≤ C for all α i w = Σ α i y i x i b= y k (1- ξ k ) - Σ α i y i x i T x k for any k s.t. α k >0 f(x) = Σ α i y i x i T x + b Again, we don’t need to compute w explicitly for classification:

13. 13 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear SVMs: Overview The classifier is a separating hyperplane. Most “important” training points are support vectors; they define the hyperplane. Quadratic optimization algorithms can identify which training points x i are support vectors with non-zero Lagrangian multipliers α i. Both in the dual formulation of the problem and in the solution training points appear only inside inner products: Find α 1 …α N such that Q( α ) = Σ α i - ½ ΣΣ α i α j y i y j x i T x j is maximized and (1) Σ α i y i = 0 (2) 0 ≤ α i ≤ C for all α i f(x) = Σ α i y i x i T x + b

14. 14 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Non-linear SVMs Datasets that are linearly separable with some noise work out great: But what are we going to do if the dataset is just too hard? How about… mapping data to a higher-dimensional space: 0 0 0 x2x2 x x x

15. 15 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Non-linear SVMs: Feature spaces General idea: the original feature space can always be mapped to some higher-dimensional feature space where the training set is separable: Φ: x → φ(x)

16. 16 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 The “Kernel Trick” The linear classifier relies on inner product between vectors K(x i,x j )=x i T x j If every datapoint is mapped into high-dimensional space via some transformation Φ: x → φ(x), the inner product becomes: K(x i,x j )= φ(x i ) T φ(x j ) A kernel function is a function that is eqiuvalent to an inner product in some feature space. Example: 2-dimensional vectors x=[x 1 x 2 ]; let K(x i,x j )=(1 + x i T x j ) 2, Need to show that K(x i,x j )= φ(x i ) T φ(x j ): K(x i,x j )=(1 + x i T x j ) 2, = 1+ x i1 2 x j1 2 + 2 x i1 x j1 x i2 x j2 + x i2 2 x j2 2 + 2x i1 x j1 + 2x i2 x j2 = = [1 x i1 2 √2 x i1 x i2 x i2 2 √2x i1 √2x i2 ] T [1 x j1 2 √2 x j1 x j2 x j2 2 √2x j1 √2x j2 ] = = φ(x i ) T φ(x j ), where φ(x) = [1 x 1 2 √2 x 1 x 2 x 2 2 √2x 1 √2x 2 ] Thus, a kernel function implicitly maps data to a high-dimensional space (without the need to compute each φ(x) explicitly).

17. 17 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 What Functions are Kernels? For some functions K(x i,x j ) checking that K(x i,x j )= φ(x i ) T φ(x j ) can be cumbersome. Mercer’s theorem: Every semi-positive definite symmetric function is a kernel Semi-positive definite symmetric functions correspond to a semi-positive definite symmetric Gram matrix: K(x1,x1)K(x1,x1)K(x1,x2)K(x1,x2)K(x1,x3)K(x1,x3)…K(x1,xn)K(x1,xn) K(x2,x1)K(x2,x1)K(x2,x2)K(x2,x2)K(x2,x3)K(x2,x3)K(x2,xn)K(x2,xn) …………… K(xn,x1)K(xn,x1)K(xn,x2)K(xn,x2)K(xn,x3)K(xn,x3)…K(xn,xn)K(xn,xn) K=

18. 18 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Examples of Kernel Functions Linear: K(x i,x j )= x i T x j –Mapping Φ: x → φ(x), where φ(x) is x itself Polynomial of power p: K(x i,x j )= (1+ x i T x j ) p –Mapping Φ: x → φ(x), where φ(x) has dimensions Gaussian (radial-basis function): K(x i,x j ) = –Mapping Φ: x → φ(x), where φ(x) is infinite-dimensional: every point is mapped to a function (a Gaussian); combination of functions for support vectors is the separator. Higher-dimensional space still has intrinsic dimensionality d (the mapping is not onto), but linear separators in it correspond to non-linear separators in original space.

19. 19 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Non-linear SVMs Mathematically Dual problem formulation: The solution is: Optimization techniques for finding α i ’s remain the same! Find α 1 …α n such that Q(α) = Σ α i - ½ ΣΣ α i α j y i y j K(x i, x j ) is maximized and (1) Σ α i y i = 0 (2) α i ≥ 0 for all α i f(x) = Σ α i y i K(x i, x j )+ b

20. 20 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 SVM applications SVMs were originally proposed by Boser, Guyon and Vapnik in 1992 and gained increasing popularity in late 1990s. SVMs are currently among the best performers for a number of classification tasks ranging from text to genomic data. SVMs can be applied to complex data types beyond feature vectors (e.g. graphs, sequences, relational data) by designing kernel functions for such data. SVM techniques have been extended to a number of tasks such as regression [Vapnik et al. ’97], principal component analysis [Schölkopf et al. ’99], etc. Most popular optimization algorithms for SVMs use decomposition to hill- climb over a subset of α i ’s at a time, e.g. SMO [Platt ’99] and [Joachims ’99] Tuning SVMs remains a black art: selecting a specific kernel and parameters is usually done in a try-and-see manner.

21. 计算机学院 计算感知 Multiple Kernel Learning

22. 22 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 The final decision function in primal

23. 23 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 A quadratic regularization on d m

24. 24 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Joint convex

25. 25 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Optimization Strategy Iteratively update the linear combination coefficient d and the dual variable (1) Fix d, update (2) Fix, update d

26. 26 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 The final decision function in dual

27. 27 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Structural SVM

28. 计算机学院 计算感知 Problem

29. 29 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Primal Formulation of Structural SVM

30. 30 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Dual Problem of Structural SVM

31. 31 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Algorithm

32. 32 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear Structural SVM

33. 计算机学院 计算感知 Structural Mutliple Kernel Learning

34. 34 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Linear combination of output functions

35. 35 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Optimization Problem

36. 36 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Convex Optimization Problem

37. 37 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Solution

38. 38 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Latent Structural SVM

39. 39 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

40. 40 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Algorithm of Latent Structural SVM Non-convex problem

41. 41 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Applications of Latent Structural SVM Object Recognition

42. 42 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Applications of Latent Structural SVM Group Activity Recognition

43. 43 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

44. 44 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Applications of Latent Structural SVM Image Annotation

45. 45 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

46. 46 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

47. 47 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

48. 48 University of Texas at Austin Machine Learning Group 计算感知 计算机学院 Applications of Latent Structural SVM Pose Estimation

49. 49 University of Texas at Austin Machine Learning Group 计算感知 计算机学院

50. 50 University of Texas at Austin Machine Learning Group 计算感知 计算机学院