Learning with primal and dual model representations

Size: px
Start display at page:

Download "Learning with primal and dual model representations"

Transcription

1 Learning with primal and dual model representations Johan Suykens KU Leuven, ESAT-STADIUS Kasteelpark Arenberg 10 B-3001 Leuven (Heverlee), Belgium CIMI 2015, Toulouse Learning with primal and dual model representations - Johan Suykens

2 Introduction and motivation Learning with primal and dual model representations - Johan Suykens

3 Data world (a) C C C 3 C 4 Learning with primal and dual model representations - Johan Suykens 1

4 Challenges data-driven general methodology scalability need for new mathematical frameworks Learning with primal and dual model representations - Johan Suykens 2

5 Different paradigms SVM & Kernel methods Convex Optimization Sparsity & Compressed sensing Learning with primal and dual model representations - Johan Suykens 3

6 Different paradigms SVM & Kernel methods Convex Optimization? Sparsity & Compressed sensing Learning with primal and dual model representations - Johan Suykens 3

7 Sparsity through regularization or loss function Learning with primal and dual model representations - Johan Suykens 3

8 Sparsity: through regularization or loss function through regularization: model ŷ = w T x + b min j w j + γ i e 2 i sparse w through loss function: model ŷ = i α ik(x, x i ) + b min w T w + γ i L(e i ) sparse α ε 0 +ε Learning with primal and dual model representations - Johan Suykens 4

9 Sparsity: matrices and tensors vector x matrix X tensor X data vector x data matrix X data tensor X vector model: matrix model: tensor model: ŷ = w T x ŷ = W,X ŷ = W, X sparsity: sparsity: sparsity: j w j W W [Signoretto M., Tran Dinh Q., De Lathauwer L., Suykens J.A.K., ML 2014] Robust tensor completion [Yang, Feng, Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 5

10 Sparsity: matrices and tensors vector x matrix X tensor X data vector x data matrix X data tensor X vector model: matrix model: tensor model: ŷ = w T x ŷ = W,X ŷ = W, X sparsity: sparsity: sparsity: j w j W W Learning with tensors [Signoretto, Tran Dinh, De Lathauwer, Suykens, ML 2014] Robust tensor completion [Yang, Feng, Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 5

11 Function estimation in RKHS Find function f such that [Wahba, 1990; Evgeniou et al., 2000] 1 min f H K N N L(y i, f(x i )) + λ f 2 K i=1 with L(, ) the loss function. f K is norm in RKHS H K defined by K. Representer theorem: for convex loss function, solution of the form f(x) = NX α i K(x, x i ) i=1 Reproducing property f(x) = f, K x K with K x ( ) = K(x, ) Sparse representation by ǫ-insensitive loss [Vapnik, 1998] Learning with primal and dual model representations - Johan Suykens 6

12 Learning with primal and dual model representations Learning with primal and dual model representations - Johan Suykens 6

13 Learning models from data: alternative views - Consider model ŷ = f(x; w), given input/output data {(x i, y i )} N i=1 : min w wt w + γ N (y i f(x i ;w)) 2 i=1 - Rewrite the problem as min w,e w T w + γ N i=1 (y i f(x i ;w)) 2 subject to e i = y i f(x i ;w), i = 1,...,N - Construct Lagrangian and take condition for optimality - For a model f(x;w) = h j=1 w jϕ j (x) = w T ϕ(x) one obtains then ˆf(x) = N i=1 α ik(x, x i ) with K(x, x i ) = ϕ(x) T ϕ(x i ). Learning with primal and dual model representations - Johan Suykens 7

14 Learning models from data: alternative views - Consider model ŷ = f(x; w), given input/output data {(x i, y i )} N i=1 : min w wt w + γ N (y i f(x i ; w)) 2 i=1 - Rewrite the problem as min w,e w T w + γ N i=1 e2 i subject to e i = y i f(x i ;w),i = 1,...,N - Express the solution and the model in terms of Lagrange multipliers α i - For a model f(x;w) = h j=1 w jϕ j (x) = w T ϕ(x) one obtains then ˆf(x) = N i=1 α ik(x, x i ) with K(x, x i ) = ϕ(x) T ϕ(x i ). Learning with primal and dual model representations - Johan Suykens 7

15 Least Squares Support Vector Machines: core models Regression Classification min w,b,e wt w + γ i min w,b,e wt w + γ i e 2 i e 2 i s.t. y i = w T ϕ(x i ) + b + e i, i s.t. y i (w T ϕ(x i ) + b) = 1 e i, i Kernel pca (V = I), Kernel spectral clustering (V = D 1 ) min w,b,e wt w + γ i v i e 2 i s.t. e i = w T ϕ(x i ) + b, i Kernel canonical correlation analysis/partial least squares min w,v,b,d,e,r wt w + v T v + ν i (e i r i ) 2 s.t. { ei = w T ϕ (1) (x i ) + b = v T ϕ (2) (y i ) + d r i [Suykens & Vandewalle, 1999; Suykens et al., 2002; Alzate & Suykens, 2010] Learning with primal and dual model representations - Johan Suykens 8

16 Probability and quantum mechanics Kernel pmf estimation Primal: 1 min w,p 2 w, w subject to p i = w,ϕ(x i ), i = 1,...,N and N i=1 p i = 1 i Dual: p i = P Nj=1 K(x j,x i ) P Ni=1 P Nj=1 K(x j,x i ) Quantum measurement: state vector ψ, measurement operators M i Primal: 1 min w,p 2 w w subject to p i = Re( w M i ψ ), i = 1,...,N and N i=1 p i = 1 i Dual: p i = ψ M i ψ (Born rule, orthogonal projective measurement) [Suykens, Physical Review A, 2013] Learning with primal and dual model representations - Johan Suykens 9

17 SVMs: living in two worlds... Primal space x x x x x o o o o Feature space ϕ(x) x x x o oo x x o Input space Parametric Dual space ŷ = sign[w T ϕ(x) + b] Nonparametric ŷ K(x i, x j ) = ϕ(x i ) T ϕ(x j ) (Mercer) ŷ = sign[ P #sv i=1 α iy i K(x, x i ) + b] ŷ w 1 w nh α 1 ϕ 1 (x) ϕ nh (x) K(x, x 1 ) x x α #sv K(x, x #sv ) Learning with primal and dual model representations - Johan Suykens 10

18 SVMs: living in two worlds... Primal space x x x x x o o o o Feature space ϕ(x) x x x o oo x x o Input space Parametric Dual space ŷ = sign[w T ϕ(x) + b] Nonparametric ŷ K(x i, x j ) = ϕ(x i ) T ϕ(x j ) ( Kernel trick ) ŷ = sign[ P #sv i=1 α iy i K(x, x i ) + b] ŷ w 1 w nh α 1 ϕ 1 (x) ϕ nh (x) K(x, x 1 ) x x Parametric Non parametric α #sv K(x, x #sv ) Learning with primal and dual model representations - Johan Suykens 10

19 Linear model: solving in primal or dual? inputs x R d, output y R training set {(x i, y i )} N i=1 Model ր ց (P) : ŷ = w T x + b, w R d (D) : ŷ = i α i x T i x + b, α RN Learning with primal and dual model representations - Johan Suykens 11

20 Linear model: solving in primal or dual? inputs x R d, output y R training set {(x i, y i )} N i=1 Model ր ց (P) : ŷ = w T x + b, w R d (D) : ŷ = i α i x T i x + b, α RN Learning with primal and dual model representations - Johan Suykens 11

21 Linear model: solving in primal or dual? few inputs, many data points: d N primal : w R d dual: α R N (large kernel matrix: N N) Learning with primal and dual model representations - Johan Suykens 12

22 Linear model: solving in primal or dual? many inputs, few data points: d N primal: w R d dual : α R N (small kernel matrix: N N) Learning with primal and dual model representations - Johan Suykens 12

23 From linear to nonlinear model: Feature map and kernel Model ր ց (P) : (D) : ŷ = w T ϕ(x) + b ŷ = i α i K(x i, x) + b Mercer theorem: K(x, z) = ϕ(x) T ϕ(z) Feature map ϕ(x) = [ϕ 1 (x); ϕ 2 (x);...;ϕ h (x)] Kernel function K(x, z) (e.g. linear, polynomial, RBF,...) Use of feature map and positive definite kernel [Cortes & Vapnik, 1995] Extension to infinite dimensional case: - LS-SVM formulation [Signoretto, De Lathauwer, Suykens, 2011] - HHK Transform, coherent states, wavelets [Fanuel & Suykens, 2015] Learning with primal and dual model representations - Johan Suykens 13

24 Coherent states { η x H} x X in HHK transform min w H,e i,b 1 2 w w H + γ 2 N i=1 e 2 i s.t. y i = η xi w H +b+e i, i = 1,...,N HHK Transform: W η : H H K : w η w H (P) : ŷ = η x w H + b transform ŷ = W η η x W η w K + b ր M K(x, z) = η x η z H K(x, z) = ξ x ξ z K, ξ x = W η η x ց (D) : ŷ = i α i K(x i,x) + b ŷ = i α i K(x i,x) + b [Fanuel & Suykens, TR15-101, 2015] Learning with primal and dual model representations - Johan Suykens 14

25 HHK transform Coherent states { η x H} x X in min w H,e i,b 1 2 w w H + γ 2 N i=1 e 2 i s.t. y i = η xi w H +b+e i, i = 1,...,N HHK Transform: W η : H H K : w η w H (P) : ŷ = η x w H + b H H K ŷ = W η η x W η w K + b ր M K(x, z) = η x η z H K(x, z) = ξ x ξ z K, ξ x = W η η x ց (D) : ŷ = i α i K(x i,x) + b ŷ = i α i K(x i, x) + b [Fanuel & Suykens, TR15-101, 2015] Learning with primal and dual model representations - Johan Suykens 14

26 Sparsity by fixed-size kernel method Learning with primal and dual model representations - Johan Suykens 14

27 Fixed-size method: steps 1. selection of a subset from the data 2. kernel matrix on the subset 3. eigenvalue decomposition of kernel matrix 4. approximation of the feature map based on the eigenvectors (Nyström approximation) 5. estimation of the model in the primal using the approximate feature map (applicable to large data sets) [Suykens et al., 2002] (ls-svm book) Learning with primal and dual model representations - Johan Suykens 15

28 Selection of subset random quadratic Renyi entropy incomplete Cholesky factorization Learning with primal and dual model representations - Johan Suykens 16

29 Nyström method big kernel matrix: Ω (N,N) R N N small kernel matrix: Ω (M,M) R M M (on subset) Eigenvalue decompositions: Ω (N,N) Ũ = Ũ Λ and Ω (M,M) U = U Λ Relation to eigenvalues and eigenfunctions of the integral equation K(x, x )φ i (x)p(x)dx = λ i φ i (x ) with ˆλ i = 1 M λ i, ˆφi (x k ) = M u ki, ˆφi (x ) = M λ i M u ki K(x k,x ) k=1 [Williams & Seeger, 2001] (Nyström method in GP) Learning with primal and dual model representations - Johan Suykens 17

30 Fixed-size method: estimation in primal For the feature map ϕ( ) : R d R h obtain an approximation ϕ( ) : R d R M based on the eigenvalue decomposition of the kernel matrix with ϕ i (x ) = ˆλi ˆφi (x ) (on a subset of size M N). Estimate in primal: min w, b 1 2 wt w + γ 1 2 N (y i w T ϕ(x i ) b) 2 i=1 Sparse representation is obtained: w R M with M N and M h. [Suykens et al., 2002; De Brabanter et al., CSDA 2010] Learning with primal and dual model representations - Johan Suykens 18

31 Fixed-size method: performance in classification pid spa mgt adu ftc N N cv N test d FS-LSSVM (# SV) C-SVM (# SV) ν-svm (# SV) RBF FS-LSSVM 76.7(3.43) 92.5(0.67) 86.6(0.51) 85.21(0.21) 81.8(0.52) Lin FS-LSSVM 77.6(0.78) 90.9(0.75) 77.8(0.23) 83.9(0.17) 75.61(0.35) RBF C-SVM 75.1(3.31) 92.6(0.76) 85.6(1.46) 84.81(0.20) 81.5(no cv) Lin C-SVM 76.1(1.76) 91.9(0.82) 77.3(0.53) 83.5(0.28) 75.24(no cv) RBF ν-svm 75.8(3.34) 88.7(0.73) 84.2(1.42) 83.9(0.23) 81.6(no cv) Maj. Rule 64.8(1.46) 60.6(0.58) 65.8(0.28) 83.4(0.1) 51.23(0.20) Fixed-size (FS) LSSVM: good performance and sparsity wrt C-SVM and ν-svm Challenging to achieve high performance by very sparse models [De Brabanter et al., CSDA 2010] Learning with primal and dual model representations - Johan Suykens 19

32 Two stages of sparsity stage 1 primal FS model estimation reweighted l 1 dual subset selection Nyström approximation Synergy between parametric & kernel-based models [Mall & Suykens, IEEE-TNNLS, 2015], reweighted l 1 [Candes et al., 2008] Other possible approaches with improved sparsity: SCAD [Fan & Li, 2001]; coefficientbased l q (0 < q 1) [Shi et al., 2013]; two-level l 1 [Huang et al., 2014] Learning with primal and dual model representations - Johan Suykens 20

33 Two stages of sparsity stage 1 primal FS model estimation reweighted l 1 dual subset selection Nyström approximation Synergy between parametric & kernel-based models [Mall & Suykens, IEEE-TNNLS, 2015], reweighted l 1 [Candes et al., 2008] Other possible approaches with improved sparsity: SCAD [Fan & Li, 2001]; coefficientbased l q (0 < q 1) [Shi et al., 2013]; two-level l 1 [Huang et al., 2014] Learning with primal and dual model representations - Johan Suykens 20

34 Two stages of sparsity stage 1 stage 2 primal FS model estimation reweighted l 1 dual subset selection Nyström approximation Synergy between parametric & kernel-based models [Mall & Suykens, IEEE-TNNLS 2015], reweighted l 1 [Candes et al., 2008] Other possible approaches with improved sparsity: SCAD [Fan & Li, 2001]; coefficientbased l q (0 < q 1) [Shi et al., 2013]; two-level l 1 [Huang et al., 2014] Learning with primal and dual model representations - Johan Suykens 20

35 Two stages of sparsity stage 1 stage 2 primal FS model estimation reweighted l 1 dual subset selection Nyström approximation Synergy between parametric & kernel-based models [Mall & Suykens, IEEE-TNNLS 2015], reweighted l 1 [Candes et al., 2008] Other possible approaches with improved sparsity: SCAD [Fan & Li, 2001]; coefficientbased l q (0 < q 1) [Shi et al., 2013]; two-level l 1 [Huang et al., 2014] Learning with primal and dual model representations - Johan Suykens 20

36 Kernel-based models for spectral clustering Learning with primal and dual model representations - Johan Suykens 20

37 Kernel PCA Primal problem: [Suykens et al., 2002] min w,b,e 1 2 wt w 1 N 2 γ i=1 e 2 i s.t. e i = w T ϕ(x i ) + b, i = 1,...,N. Dual problem corresponds to kernel PCA [Scholkopf et al., 1998] Ω c α = λα with λ = 1/γ with Ω c,ij = (ϕ(x i ) ˆµ ϕ ) T (ϕ(x j ) ˆµ ϕ ) the centered kernel matrix. Interpretation: 1. pool of candidate components (objective function equals zero) 2. select relevant components Robust and sparse versions [Alzate & Suykens, 2008]: by taking other loss functions Learning with primal and dual model representations - Johan Suykens 21

38 Robustness: Kernel Component Analysis original image corrupted image KPCA reconstruction KCA reconstruction Weighted LS-SVM [Alzate & Suykens, IEEE-TNN 2008]: robustness and sparsity Learning with primal and dual model representations - Johan Suykens 22

39 Kernel Spectral Clustering (KSC): case of two clusters Primal problem: training on given data {x i } N i=1 1 min w,b,e 2 wt w γ 1 2 et V e subject to e i = w T ϕ(x i ) + b, i = 1,...,N with weighting matrix V and ϕ( ) : R d R h the feature map. Dual: V M V Ωα = λα with λ = 1/γ, M V = I N T N V 1 N 1 T NV weighted centering matrix, N Ω = [Ω ij ] kernel matrix with Ω ij = ϕ(x i ) T ϕ(x j ) = K(x i, x j ) Taking V = D 1 with degree matrix D = diag{d 1,...,d N } and d i = N j=1 Ω ij relates to random walks algorithm. [Alzate & Suykens, IEEE-PAMI, 2010] Learning with primal and dual model representations - Johan Suykens 23

40 Lagrangian: Lagrangian and conditions for optimality L(w,b, e;α) = 1 2 wt w γ 1 2 N v i e 2 i + i=1 N α i (e i w T ϕ(x i ) b) i=1 Conditions for optimality: L w = 0 w = i α iϕ(x i ) L b = 0 i α i = 0 L = 0 α i = γv i e i, i = 1,...,N e i L = 0 e i = w T ϕ(x i ) + b, i = 1,...,N α i Eliminate w,b,e, write solution in Lagrange multipliers α i. Learning with primal and dual model representations - Johan Suykens 24

41 Kernel spectral clustering: more clusters Case of k clusters: additional sets of constraints 1 min w (l),e (l),b l 2 k 1 l=1 w (l)t w (l) 1 2 k 1 l=1 γ l e (l)t D 1 e (l) subject to e (1) = Φ N nh w (1) + b 1 1 N e (2) = Φ N nh w (2) + b 2 1 N. e (k 1) = Φ N nh w (k 1) + b k 1 1 N where e (l) = [e (l) 1 ;...;e(l) N ] and Φ N n h = [ϕ(x 1 ) T ;...;ϕ(x N ) T ] R N n h. Dual problem: M D Ωα (l) = λdα (l), l = 1,...,k 1. [Alzate & Suykens, IEEE-PAMI, 2010] Learning with primal and dual model representations - Johan Suykens 25

42 Primal and dual model representations k clusters k 1 sets of constraints (index l = 1,...,k 1) M ր ց (P) : sign[ê (l) ] = sign[w (l)t ϕ(x ) + b l ] (D) : sign[ê (l) ] = sign[ j α(l) j K(x,x j ) + b l ] Learning with primal and dual model representations - Johan Suykens 26

43 Advantages of kernel-based setting model-based approach out-of-sample extensions, applying model to new data consider training, validation and test data (training problem corresponds to eigenvalue decomposition problem) model selection procedures sparse representations and large scale methods Learning with primal and dual model representations - Johan Suykens 27

44 Model selection: toy example e (2) i,val σ 2 = 0.5, BLF = e (2) i,val e (1) i,val σ 2 = 0.16, BLF = e (1) i,val validation set x (2) x (2) x (1) x (1) train + validation + test data BAD GOOD Learning with primal and dual model representations - Johan Suykens 28

45 Example: image segmentation i,val e (3) e (2) i,val e (1) i,val Learning with primal and dual model representations - Johan Suykens 29

46 Hierarchical KSC [Alzate & Suykens, 2012] Learning with primal and dual model representations - Johan Suykens 30

47 Hierarchical KSC [Alzate & Suykens, 2012] Learning with primal and dual model representations - Johan Suykens 31

48 Kernel spectral clustering: sparse kernel models original image binary clustering Incomplete Cholesky decomposition: Ω GG T with G R N R and R N Image (Berkeley image dataset): (154, 401 pixels), 175 SV Time-complexity O(R 2 N 2 ) in [Alzate & Suykens, 2008] Time-complexity O(R 2 N) in [Novak, Alzate, Langone, Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 32

49 Kernel spectral clustering: sparse kernel models original image sparse kernel model Incomplete Cholesky decomposition: Ω GG T with G R N R and R N Image (Berkeley image dataset): (154, 401 pixels), 175 SV Time-complexity O(R 2 N 2 ) in [Alzate & Suykens, 2008] Time-complexity O(R 2 N) in [Novak, Alzate, Langone, Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 32

50 Incomplete Cholesky decomposition and reduced set For KSC problem M D Ωα = λdα, solve the approximation U T M D UΛ 2 ζ = λζ from Ω GG T, singular value decomposition G = UΛV T and ζ = U T α. A smaller matrix of size R R is obtained instead of N N. Pivots are used as subset { x i } for the data Reduced set method [Scholkopf et al., 1999]: approximation of w = N i=1 α iϕ(x i ) by w = M j=1 β jϕ( x j ) in the sense min w w 2 2 β j β j Sparser solutions by adding l 1 penalty, reweighted l 1 or group Lasso. [Alzate & Suykens, 2008, 2011; Mall & Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 33

51 Incomplete Cholesky decomposition and reduced set For KSC problem M D Ωα = λdα, solve the approximation U T M D UΛ 2 ζ = λζ from Ω GG T, singular value decomposition G = UΛV T and ζ = U T α. A smaller matrix of size R R is obtained instead of N N. Pivots are used as subset { x i } for the data Reduced set method [Scholkopf et al., 1999]: approximation of w = N i=1 α iϕ(x i ) by w = M j=1 β jϕ( x j ) in the sense min β w w ν j β j Sparser solutions by adding l 1 penalty, reweighted l 1 or group Lasso. [Alzate & Suykens, 2008, 2011; Mall & Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 33

52 Core models + constraints Core model + regularization terms additional constraints Learning with primal and dual model representations - Johan Suykens 34

53 Core models + constraints Core model + regularization terms additional constraints optimal model representation model estimate Learning with primal and dual model representations - Johan Suykens 34

54 Kernel spectral clustering: adding prior knowledge Pair of points x, x : c = 1 must-link, c = 1 cannot-link Primal problem [Alzate & Suykens, IJCNN 2009] min w (l),e (l),b l 1 2 k 1 l=1 w (l)t w (l) k 1 l=1 γ l e (l)t D 1 e (l) subject to e (1) = Φ N nh w (1) + b 1 1 N. e (k 1) = Φ N nh w (k 1) + b k 1 1 N w (1)T ϕ(x ) = cw (1)T ϕ(x ). w (k 1)T ϕ(x ) = cw (k 1)T ϕ(x ) Dual problem: yields rank-one downdate of the kernel matrix Learning with primal and dual model representations - Johan Suykens 35

55 Adding prior knowledge original image without constraints Learning with primal and dual model representations - Johan Suykens 36

56 Adding prior knowledge original image with constraints Learning with primal and dual model representations - Johan Suykens 36

57 Semi-supervised learning using KSC (1) N unlabeled data, but additional labels on M N data X = {x 1,...,x N,x N+1,...,x M } Kernel spectral clustering as core model (binary case [Alzate & Suykens, WCCI 2012], multi-way/multi-class [Mehrkanoon et al., TNNLS 2015]) min w,e,b 1 2 wt w γ 1 2 et D 1 e+ρ 1 2 M m=n+1 subject to e i = w T ϕ(x i ) + b, i = 1,...,M (e m y m ) 2 Dual solution is characterized by a linear system. Suitable for clustering as well as classification. Other approaches in semi-supervised learning and manifold learning, e.g. [Belkin et al., 2006] Learning with primal and dual model representations - Johan Suykens 37

58 Semi-supervised learning using KSC (2) Dataset size n L /n U test (%) FS semi-ksc RD semi-ksc Lap-SVMp Spambase / (20%) ± ± ± 0.03 Satimage / (20%) ± ± ± Ring / (20%) ± ± ± Magic / (20%) ± ± ± Cod-rna / (20%) ± ± ± Covertype / (5%) ± ± ± / ± ± / ± ± / ± ± 0.06 FS semi-ksc: Fixed-size semi-supervised KSC RD semi-ksc: other subset selection related to [Lee & Mangasarian, 2001] Lap-SVM: Laplacian support vector machine [Belkin et al., 2006] [Mehrkanoon & Suykens, 2014] Learning with primal and dual model representations - Johan Suykens 38

59 Semi-supervised learning using KSC (3) original image KSC given a few labels semi-supervised KSC [Mehrkanoon, Alzate, Mall, Langone, Suykens, IEEE-TNNLS 2015], videos Learning with primal and dual model representations - Johan Suykens 39

60 SVD from LS-SVM Learning with primal and dual model representations - Johan Suykens 39

61 SVD within the LS-SVM setting (1) Singular Value Decomposition (SVD) of A R N M A = UΣV T with U T U = I N, V T V = I M, Σ = diag(σ 1,...,σ p ) R N M. Obtain two sets of data points (rows and columns): x i = A T ǫ i, z j = Aε j for i = 1,...,N, j = 1,...,M where ǫ i, ε j are standard basis vectors of dimension N and M. Compatible feature maps: ϕ : R M R N, ψ : R N R N where ϕ(x i ) = C T x i = C T A T ǫ i ψ(z j ) = z j = Aε j with C R M N a compatibility matrix. [Suykens, ACHA, 2015, in press] Learning with primal and dual model representations - Johan Suykens 40

62 Primal problem: SVD within the LS-SVM setting (2) min w,v,e,r wt v γ N i=1 e2 i γ M j=1 r2 j subject to e i = w T ϕ(x i ), i = 1,...,N r j = v T ψ(z j ), j = 1,...,M From the Lagrangian and conditions for optimality one obtains: [ ] ϕ(x i ) T ψ(z j ) [β] = [α] Λ [ ] ψ(z j ) T ϕ(x i ) [α] = [β] Λ Theorem: If ACA = A holds, this corresponds to the shifted eigenvalue problem in Lanczos decomposition theorem. Goes beyond the use of Mercer theorem; extensions to nonlinear SVDs [Suykens, ACHA, 2015, in press] Learning with primal and dual model representations - Johan Suykens 41

63 Conclusions Synergies parametric and kernel based-modelling Primal and dual representations Sparse kernel models using fixed-size method Applications in supervised and unsupervised learning and beyond Finite and infinite dimensional case Beyond Mercer kernels Software: see ERC AdG A-DATADRIVE-B website Learning with primal and dual model representations - Johan Suykens 42

64 Acknowledgements (1) Co-workers at ESAT-STADIUS: M. Agudelo, C. Alaiz, C. Alzate, A. Argyriou, R. Castro, J. De Brabanter, K. De Brabanter, L. De Lathauwer, B. De Moor, M. Espinoza, M. Fanuel, Y. Feng, E. Frandi, B. Gauthier, D. Geebelen, H. Hang, X. Huang, L. Houthuys, V. Jumutc, Z. Karevan, R. Langone, Y. Liu, R. Mall, S. Mehrkanoon, M. Novak, J. Puertas, L. Shi, M. Signoretto, V. Van Belle, J. Vandewalle, S. Van Huffel, C. Varon, X. Xi, Y. Yang, and others Many people for joint work, discussions, invitations, organizations Support from ERC AdG A-DATADRIVE-B, KU Leuven, GOA-MaNet, OPTEC, IUAP DYSCO, FWO projects, IWT, iminds, BIL, COST Learning with primal and dual model representations - Johan Suykens 43

65 Acknowledgements (2) Learning with primal and dual model representations - Johan Suykens 44

66 Thank you Learning with primal and dual model representations - Johan Suykens 45

Learning with primal and dual model representations: a unifying picture

Learning with primal and dual model representations: a unifying picture Learning with primal and dual model representations: a unifying picture Johan Suykens KU Leuven, ESAT-STADIUS Kasteelpark Arenberg 10 B-3001 Leuven (Heverlee), Belgium Email: johan.suykens@esat.kuleuven.be

More information

Spatio-temporal feature selection for black-box weather forecasting

Spatio-temporal feature selection for black-box weather forecasting Spatio-temporal feature selection for black-box weather forecasting Zahra Karevan and Johan A. K. Suykens KU Leuven, ESAT-STADIUS Kasteelpark Arenberg 10, B-3001 Leuven, Belgium Abstract. In this paper,

More information

Support vector machines and kernel-based learning for dynamical systems modelling

Support vector machines and kernel-based learning for dynamical systems modelling Support vector machines and kernel-based learning for dynamical systems modelling Johan Suykens K.U. Leuven, ESAT-SCD/SISTA Kasteelpark Arenberg 0 B-300 Leuven (Heverlee), Belgium Email: johan.suykens@esat.kuleuven.be

More information

Primal and Dual Model Representations in Kernel-based Learning

Primal and Dual Model Representations in Kernel-based Learning Primal and Dual Model Representations in Kernel-based Learning Johan Suykens K.U. Leuven, ESAT-SCD/SISTA Kasteelpark Arenberg B- Leuven (Heverlee), Belgium Email: johan.suykens@esat.kuleuven.be http://www.esat.kuleuven.be/scd/

More information

Support Vector Machines and Kernel Based Learning

Support Vector Machines and Kernel Based Learning Support Vector Machines and Kernel Based Learning Johan Suykens K.U. Leuven, ESAT-SCD/SISTA Kasteelpark Arenberg B-3 Leuven (Heverlee), Belgium Tel: 32/6/32 8 2 - Fa: 32/6/32 9 7 Email: johan.suykens@esat.kuleuven.be

More information

Support Vector Machines for Classification: A Statistical Portrait

Support Vector Machines for Classification: A Statistical Portrait Support Vector Machines for Classification: A Statistical Portrait Yoonkyung Lee Department of Statistics The Ohio State University May 27, 2011 The Spring Conference of Korean Statistical Society KAIST,

More information

Perceptron Revisited: Linear Separators. Support Vector Machines

Perceptron Revisited: Linear Separators. Support Vector Machines Support Vector Machines 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 Department

More information

Spatio-temporal feature selection for black-box weather forecasting

Spatio-temporal feature selection for black-box weather forecasting ESANN 06 proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges (Belgium), 7-9 April 06, i6doc.com publ., ISBN 978-8758707-8. Spatio-temporal

More information

Compactly supported RBF kernels for sparsifying the Gram matrix in LS-SVM regression models

Compactly supported RBF kernels for sparsifying the Gram matrix in LS-SVM regression models Compactly supported RBF kernels for sparsifying the Gram matrix in LS-SVM regression models B. Hamers, J.A.K. Suykens, B. De Moor K.U.Leuven, ESAT-SCD/SISTA, Kasteelpark Arenberg, B-3 Leuven, Belgium {bart.hamers,johan.suykens}@esat.kuleuven.ac.be

More information

Least Squares SVM Regression

Least Squares SVM Regression Least Squares SVM Regression Consider changing SVM to LS SVM by making following modifications: min (w,e) ½ w 2 + ½C Σ e(i) 2 subject to d(i) (w T Φ( x(i))+ b) = e(i), i, and C>0. Note that e(i) is error

More information

EE613 Machine Learning for Engineers. Kernel methods Support Vector Machines. jean-marc odobez 2015

EE613 Machine Learning for Engineers. Kernel methods Support Vector Machines. jean-marc odobez 2015 EE613 Machine Learning for Engineers Kernel methods Support Vector Machines jean-marc odobez 2015 overview Kernel methods introductions and main elements defining kernels Kernelization of k-nn, K-Means,

More information

Learning with matrix and tensor based models using low-rank penalties

Learning with matrix and tensor based models using low-rank penalties Learning with matrix and tensor based models using low-rank penalties Johan Suykens KU Leuven, ESAT-SCD/SISTA Kasteelpark Arenberg 10 B-3001 Leuven (Heverlee), Belgium Email: johan.suykens@esat.kuleuven.be

More information

Accepted Manuscript. Indefinite kernels in least squares support vector machines and principal component analysis

Accepted Manuscript. Indefinite kernels in least squares support vector machines and principal component analysis Accepted Manuscript Indefinite kernels in least squares support vector machines and principal component analysis Xiaolin Huang, Andreas Maier, Joachim Hornegger, Johan A.K. Suykens PII: S63-523(6)349-5

More information

Introduction to Support Vector Machines

Introduction to Support Vector Machines Introduction to Support Vector Machines Shivani Agarwal Support Vector Machines (SVMs) Algorithm for learning linear classifiers Motivated by idea of maximizing margin Efficient extension to non-linear

More information

Basis Expansion and Nonlinear SVM. Kai Yu

Basis Expansion and Nonlinear SVM. Kai Yu Basis Expansion and Nonlinear SVM Kai Yu Linear Classifiers f(x) =w > x + b z(x) = sign(f(x)) Help to learn more general cases, e.g., nonlinear models 8/7/12 2 Nonlinear Classifiers via Basis Expansion

More information

Outline. Basic concepts: SVM and kernels SVM primal/dual problems. Chih-Jen Lin (National Taiwan Univ.) 1 / 22

Outline. Basic concepts: SVM and kernels SVM primal/dual problems. Chih-Jen Lin (National Taiwan Univ.) 1 / 22 Outline Basic concepts: SVM and kernels SVM primal/dual problems Chih-Jen Lin (National Taiwan Univ.) 1 / 22 Outline Basic concepts: SVM and kernels Basic concepts: SVM and kernels SVM primal/dual problems

More information

Approximate Kernel PCA with Random Features

Approximate Kernel PCA with Random Features Approximate Kernel PCA with Random Features (Computational vs. Statistical Tradeoff) Bharath K. Sriperumbudur Department of Statistics, Pennsylvania State University Journées de Statistique Paris May 28,

More information

Approximate Kernel Methods

Approximate Kernel Methods Lecture 3 Approximate Kernel Methods Bharath K. Sriperumbudur Department of Statistics, Pennsylvania State University Machine Learning Summer School Tübingen, 207 Outline Motivating example Ridge regression

More information

CIS 520: Machine Learning Oct 09, Kernel Methods

CIS 520: Machine Learning Oct 09, Kernel Methods CIS 520: Machine Learning Oct 09, 207 Kernel Methods Lecturer: Shivani Agarwal Disclaimer: These notes are designed to be a supplement to the lecture They may or may not cover all the material discussed

More information

Jeff Howbert Introduction to Machine Learning Winter

Jeff Howbert Introduction to Machine Learning Winter Classification / Regression Support Vector Machines Jeff Howbert Introduction to Machine Learning Winter 2012 1 Topics SVM classifiers for linearly separable classes SVM classifiers for non-linearly separable

More information

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers)

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Support vector machines In a nutshell Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Solution only depends on a small subset of training

More information

Machine Learning. Support Vector Machines. Manfred Huber

Machine Learning. Support Vector Machines. Manfred Huber Machine Learning Support Vector Machines Manfred Huber 2015 1 Support Vector Machines Both logistic regression and linear discriminant analysis learn a linear discriminant function to separate the data

More information

Review: Support vector machines. Machine learning techniques and image analysis

Review: Support vector machines. Machine learning techniques and image analysis Review: Support vector machines Review: Support vector machines Margin optimization min (w,w 0 ) 1 2 w 2 subject to y i (w 0 + w T x i ) 1 0, i = 1,..., n. Review: Support vector machines Margin optimization

More information

Kernel Methods. Barnabás Póczos

Kernel Methods. Barnabás Póczos Kernel Methods Barnabás Póczos Outline Quick Introduction Feature space Perceptron in the feature space Kernels Mercer s theorem Finite domain Arbitrary domain Kernel families Constructing new kernels

More information

An Improved Conjugate Gradient Scheme to the Solution of Least Squares SVM

An Improved Conjugate Gradient Scheme to the Solution of Least Squares SVM An Improved Conjugate Gradient Scheme to the Solution of Least Squares SVM Wei Chu Chong Jin Ong chuwei@gatsby.ucl.ac.uk mpeongcj@nus.edu.sg S. Sathiya Keerthi mpessk@nus.edu.sg Control Division, Department

More information

LMS Algorithm Summary

LMS Algorithm Summary LMS Algorithm Summary Step size tradeoff Other Iterative Algorithms LMS algorithm with variable step size: w(k+1) = w(k) + µ(k)e(k)x(k) When step size µ(k) = µ/k algorithm converges almost surely to optimal

More information

Support'Vector'Machines. Machine(Learning(Spring(2018 March(5(2018 Kasthuri Kannan

Support'Vector'Machines. Machine(Learning(Spring(2018 March(5(2018 Kasthuri Kannan Support'Vector'Machines Machine(Learning(Spring(2018 March(5(2018 Kasthuri Kannan kasthuri.kannan@nyumc.org Overview Support Vector Machines for Classification Linear Discrimination Nonlinear Discrimination

More information

Indirect Rule Learning: Support Vector Machines. Donglin Zeng, Department of Biostatistics, University of North Carolina

Indirect Rule Learning: Support Vector Machines. Donglin Zeng, Department of Biostatistics, University of North Carolina Indirect Rule Learning: Support Vector Machines Indirect learning: loss optimization It doesn t estimate the prediction rule f (x) directly, since most loss functions do not have explicit optimizers. Indirection

More information

Lecture 10: A brief introduction to Support Vector Machine

Lecture 10: A brief introduction to Support Vector Machine Lecture 10: A brief introduction to Support Vector Machine Advanced Applied Multivariate Analysis STAT 2221, Fall 2013 Sungkyu Jung Department of Statistics, University of Pittsburgh Xingye Qiao Department

More information

Support Vector Machine (continued)

Support Vector Machine (continued) Support Vector Machine continued) Overlapping class distribution: In practice the class-conditional distributions may overlap, so that the training data points are no longer linearly separable. We need

More information

A Least Squares Formulation for Canonical Correlation Analysis

A Least Squares Formulation for Canonical Correlation Analysis A Least Squares Formulation for Canonical Correlation Analysis Liang Sun, Shuiwang Ji, and Jieping Ye Department of Computer Science and Engineering Arizona State University Motivation Canonical Correlation

More information

Foundation of Intelligent Systems, Part I. SVM s & Kernel Methods

Foundation of Intelligent Systems, Part I. SVM s & Kernel Methods Foundation of Intelligent Systems, Part I SVM s & Kernel Methods mcuturi@i.kyoto-u.ac.jp FIS - 2013 1 Support Vector Machines The linearly-separable case FIS - 2013 2 A criterion to select a linear classifier:

More information

Linear & nonlinear classifiers

Linear & nonlinear classifiers Linear & nonlinear classifiers Machine Learning Hamid Beigy Sharif University of Technology Fall 1394 Hamid Beigy (Sharif University of Technology) Linear & nonlinear classifiers Fall 1394 1 / 34 Table

More information

Statistical Machine Learning from Data

Statistical Machine Learning from Data Samy Bengio Statistical Machine Learning from Data 1 Statistical Machine Learning from Data Support Vector Machines Samy Bengio IDIAP Research Institute, Martigny, Switzerland, and Ecole Polytechnique

More information

Joint Regression and Linear Combination of Time Series for Optimal Prediction

Joint Regression and Linear Combination of Time Series for Optimal Prediction Joint Regression and Linear Combination of Time Series for Optimal Prediction Dries Geebelen 1, Kim Batselier 1, Philippe Dreesen 1, Marco Signoretto 1, Johan Suykens 1, Bart De Moor 1, Joos Vandewalle

More information

Complex networks, synchronization and cooperative behaviour

Complex networks, synchronization and cooperative behaviour Complex networks, synchronization and cooperative behaviour Johan Suykens KU Leuven, ESAT-SCD/SISTA Kasteelpark Arenberg 10 B-3001 Leuven (Heverlee), Belgium Email: johan.suykens@esat.kuleuven.be http://www.esat.kuleuven.be/scd/

More information

Chapter 9. Support Vector Machine. Yongdai Kim Seoul National University

Chapter 9. Support Vector Machine. Yongdai Kim Seoul National University Chapter 9. Support Vector Machine Yongdai Kim Seoul National University 1. Introduction Support Vector Machine (SVM) is a classification method developed by Vapnik (1996). It is thought that SVM improved

More information

Ramp Loss Linear Programming Support Vector Machine

Ramp Loss Linear Programming Support Vector Machine Journal of Machine Learning Research 5 (204) 285-22 Submitted 7/3; Revised /3; Published 6/4 Ramp Loss Linear Programming Support Vector Machine Xiaolin Huang huangxl06@mails.tsinghua.edu.cn Department

More information

Support Vector Machine (SVM) and Kernel Methods

Support Vector Machine (SVM) and Kernel Methods Support Vector Machine (SVM) and Kernel Methods CE-717: Machine Learning Sharif University of Technology Fall 2014 Soleymani Outline Margin concept Hard-Margin SVM Soft-Margin SVM Dual Problems of Hard-Margin

More information

Kernel Methods. Machine Learning A W VO

Kernel Methods. Machine Learning A W VO Kernel Methods Machine Learning A 708.063 07W VO Outline 1. Dual representation 2. The kernel concept 3. Properties of kernels 4. Examples of kernel machines Kernel PCA Support vector regression (Relevance

More information

Support Vector Machine & Its Applications

Support Vector Machine & Its Applications Support Vector Machine & Its Applications A portion (1/3) of the slides are taken from Prof. Andrew Moore s SVM tutorial at http://www.cs.cmu.edu/~awm/tutorials Mingyue Tan The University of British Columbia

More information

Support Vector Machines

Support Vector Machines Wien, June, 2010 Paul Hofmarcher, Stefan Theussl, WU Wien Hofmarcher/Theussl SVM 1/21 Linear Separable Separating Hyperplanes Non-Linear Separable Soft-Margin Hyperplanes Hofmarcher/Theussl SVM 2/21 (SVM)

More information

Kernel Principal Component Analysis

Kernel Principal Component Analysis Kernel Principal Component Analysis Seungjin Choi Department of Computer Science and Engineering Pohang University of Science and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjin@postech.ac.kr

More information

Nonlinear Dimensionality Reduction

Nonlinear Dimensionality Reduction Outline Hong Chang Institute of Computing Technology, Chinese Academy of Sciences Machine Learning Methods (Fall 2012) Outline Outline I 1 Kernel PCA 2 Isomap 3 Locally Linear Embedding 4 Laplacian Eigenmap

More information

A GENERAL FORMULATION FOR SUPPORT VECTOR MACHINES. Wei Chu, S. Sathiya Keerthi, Chong Jin Ong

A GENERAL FORMULATION FOR SUPPORT VECTOR MACHINES. Wei Chu, S. Sathiya Keerthi, Chong Jin Ong A GENERAL FORMULATION FOR SUPPORT VECTOR MACHINES Wei Chu, S. Sathiya Keerthi, Chong Jin Ong Control Division, Department of Mechanical Engineering, National University of Singapore 0 Kent Ridge Crescent,

More information

Connection of Local Linear Embedding, ISOMAP, and Kernel Principal Component Analysis

Connection of Local Linear Embedding, ISOMAP, and Kernel Principal Component Analysis Connection of Local Linear Embedding, ISOMAP, and Kernel Principal Component Analysis Alvina Goh Vision Reading Group 13 October 2005 Connection of Local Linear Embedding, ISOMAP, and Kernel Principal

More information

Structured Matrices and Solving Multivariate Polynomial Equations

Structured Matrices and Solving Multivariate Polynomial Equations Structured Matrices and Solving Multivariate Polynomial Equations Philippe Dreesen Kim Batselier Bart De Moor KU Leuven ESAT/SCD, Kasteelpark Arenberg 10, B-3001 Leuven, Belgium. Structured Matrix Days,

More information

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning Kernel Machines Pradeep Ravikumar Co-instructor: Manuela Veloso Machine Learning 10-701 SVM linearly separable case n training points (x 1,, x n ) d features x j is a d-dimensional vector Primal problem:

More information

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers)

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Support vector machines In a nutshell Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Solution only depends on a small subset of training

More information

Introduction to Machine Learning. PCA and Spectral Clustering. Introduction to Machine Learning, Slides: Eran Halperin

Introduction to Machine Learning. PCA and Spectral Clustering. Introduction to Machine Learning, Slides: Eran Halperin 1 Introduction to Machine Learning PCA and Spectral Clustering Introduction to Machine Learning, 2013-14 Slides: Eran Halperin Singular Value Decomposition (SVD) The singular value decomposition (SVD)

More information

Data Mining. Linear & nonlinear classifiers. Hamid Beigy. Sharif University of Technology. Fall 1396

Data Mining. Linear & nonlinear classifiers. Hamid Beigy. Sharif University of Technology. Fall 1396 Data Mining Linear & nonlinear classifiers Hamid Beigy Sharif University of Technology Fall 1396 Hamid Beigy (Sharif University of Technology) Data Mining Fall 1396 1 / 31 Table of contents 1 Introduction

More information

(Kernels +) Support Vector Machines

(Kernels +) Support Vector Machines (Kernels +) Support Vector Machines Machine Learning Torsten Möller Reading Chapter 5 of Machine Learning An Algorithmic Perspective by Marsland Chapter 6+7 of Pattern Recognition and Machine Learning

More information

Lecture 18: Kernels Risk and Loss Support Vector Regression. Aykut Erdem December 2016 Hacettepe University

Lecture 18: Kernels Risk and Loss Support Vector Regression. Aykut Erdem December 2016 Hacettepe University Lecture 18: Kernels Risk and Loss Support Vector Regression Aykut Erdem December 2016 Hacettepe University Administrative We will have a make-up lecture on next Saturday December 24, 2016 Presentations

More information

Linear & nonlinear classifiers

Linear & nonlinear classifiers Linear & nonlinear classifiers Machine Learning Hamid Beigy Sharif University of Technology Fall 1396 Hamid Beigy (Sharif University of Technology) Linear & nonlinear classifiers Fall 1396 1 / 44 Table

More information

Discriminative Learning and Big Data

Discriminative Learning and Big Data AIMS-CDT Michaelmas 2016 Discriminative Learning and Big Data Lecture 2: Other loss functions and ANN Andrew Zisserman Visual Geometry Group University of Oxford http://www.robots.ox.ac.uk/~vgg Lecture

More information

Kernel Methods and Support Vector Machines

Kernel Methods and Support Vector Machines Kernel Methods and Support Vector Machines Oliver Schulte - CMPT 726 Bishop PRML Ch. 6 Support Vector Machines Defining Characteristics Like logistic regression, good for continuous input features, discrete

More information

Linear Classification and SVM. Dr. Xin Zhang

Linear Classification and SVM. Dr. Xin Zhang Linear Classification and SVM Dr. Xin Zhang Email: eexinzhang@scut.edu.cn What is linear classification? Classification is intrinsically non-linear It puts non-identical things in the same class, so a

More information

The Laplacian PDF Distance: A Cost Function for Clustering in a Kernel Feature Space

The Laplacian PDF Distance: A Cost Function for Clustering in a Kernel Feature Space The Laplacian PDF Distance: A Cost Function for Clustering in a Kernel Feature Space Robert Jenssen, Deniz Erdogmus 2, Jose Principe 2, Torbjørn Eltoft Department of Physics, University of Tromsø, Norway

More information

Short Course Robust Optimization and Machine Learning. 3. Optimization in Supervised Learning

Short Course Robust Optimization and Machine Learning. 3. Optimization in Supervised Learning Short Course Robust Optimization and 3. Optimization in Supervised EECS and IEOR Departments UC Berkeley Spring seminar TRANSP-OR, Zinal, Jan. 16-19, 2012 Outline Overview of Supervised models and variants

More information

Advanced Introduction to Machine Learning

Advanced Introduction to Machine Learning 10-715 Advanced Introduction to Machine Learning Homework Due Oct 15, 10.30 am Rules Please follow these guidelines. Failure to do so, will result in loss of credit. 1. Homework is due on the due date

More information

Diffeomorphic Warping. Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel)

Diffeomorphic Warping. Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel) Diffeomorphic Warping Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel) What Manifold Learning Isn t Common features of Manifold Learning Algorithms: 1-1 charting Dense sampling Geometric Assumptions

More information

Lecture 7: Kernels for Classification and Regression

Lecture 7: Kernels for Classification and Regression Lecture 7: Kernels for Classification and Regression CS 194-10, Fall 2011 Laurent El Ghaoui EECS Department UC Berkeley September 15, 2011 Outline Outline A linear regression problem Linear auto-regressive

More information

Support Vector Machines II. CAP 5610: Machine Learning Instructor: Guo-Jun QI

Support Vector Machines II. CAP 5610: Machine Learning Instructor: Guo-Jun QI Support Vector Machines II CAP 5610: Machine Learning Instructor: Guo-Jun QI 1 Outline Linear SVM hard margin Linear SVM soft margin Non-linear SVM Application Linear Support Vector Machine An optimization

More information

References. Lecture 7: Support Vector Machines. Optimum Margin Perceptron. Perceptron Learning Rule

References. Lecture 7: Support Vector Machines. Optimum Margin Perceptron. Perceptron Learning Rule References Lecture 7: Support Vector Machines Isabelle Guyon guyoni@inf.ethz.ch An training algorithm for optimal margin classifiers Boser-Guyon-Vapnik, COLT, 992 http://www.clopinet.com/isabelle/p apers/colt92.ps.z

More information

Support Vector Machines

Support Vector Machines Support Vector Machines Sridhar Mahadevan mahadeva@cs.umass.edu University of Massachusetts Sridhar Mahadevan: CMPSCI 689 p. 1/32 Margin Classifiers margin b = 0 Sridhar Mahadevan: CMPSCI 689 p.

More information

A Spectral Regularization Framework for Multi-Task Structure Learning

A Spectral Regularization Framework for Multi-Task Structure Learning A Spectral Regularization Framework for Multi-Task Structure Learning Massimiliano Pontil Department of Computer Science University College London (Joint work with A. Argyriou, T. Evgeniou, C.A. Micchelli,

More information

Support Vector Machine (SVM) and Kernel Methods

Support Vector Machine (SVM) and Kernel Methods Support Vector Machine (SVM) and Kernel Methods CE-717: Machine Learning Sharif University of Technology Fall 2016 Soleymani Outline Margin concept Hard-Margin SVM Soft-Margin SVM Dual Problems of Hard-Margin

More information

Support Vector Machines: Maximum Margin Classifiers

Support Vector Machines: Maximum Margin Classifiers Support Vector Machines: Maximum Margin Classifiers Machine Learning and Pattern Recognition: September 16, 2008 Piotr Mirowski Based on slides by Sumit Chopra and Fu-Jie Huang 1 Outline What is behind

More information

Support Vector Machine (SVM) and Kernel Methods

Support Vector Machine (SVM) and Kernel Methods Support Vector Machine (SVM) and Kernel Methods CE-717: Machine Learning Sharif University of Technology Fall 2015 Soleymani Outline Margin concept Hard-Margin SVM Soft-Margin SVM Dual Problems of Hard-Margin

More information

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers)

Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Support vector machines In a nutshell Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers) Solution only depends on a small subset of training

More information

Support Vector Machine (SVM) & Kernel CE-717: Machine Learning Sharif University of Technology. M. Soleymani Fall 2012

Support Vector Machine (SVM) & Kernel CE-717: Machine Learning Sharif University of Technology. M. Soleymani Fall 2012 Support Vector Machine (SVM) & Kernel CE-717: Machine Learning Sharif University of Technology M. Soleymani Fall 2012 Linear classifier Which classifier? x 2 x 1 2 Linear classifier Margin concept x 2

More information

Lecture 10: Support Vector Machine and Large Margin Classifier

Lecture 10: Support Vector Machine and Large Margin Classifier Lecture 10: Support Vector Machine and Large Margin Classifier Applied Multivariate Analysis Math 570, Fall 2014 Xingye Qiao Department of Mathematical Sciences Binghamton University E-mail: qiao@math.binghamton.edu

More information

Less is More: Computational Regularization by Subsampling

Less is More: Computational Regularization by Subsampling Less is More: Computational Regularization by Subsampling Lorenzo Rosasco University of Genova - Istituto Italiano di Tecnologia Massachusetts Institute of Technology lcsl.mit.edu joint work with Alessandro

More information

Learning gradients: prescriptive models

Learning gradients: prescriptive models Department of Statistical Science Institute for Genome Sciences & Policy Department of Computer Science Duke University May 11, 2007 Relevant papers Learning Coordinate Covariances via Gradients. Sayan

More information

Training Support Vector Machines: Status and Challenges

Training Support Vector Machines: Status and Challenges ICML Workshop on Large Scale Learning Challenge July 9, 2008 Chih-Jen Lin (National Taiwan Univ.) 1 / 34 Training Support Vector Machines: Status and Challenges Chih-Jen Lin Department of Computer Science

More information

Introduction to Gaussian Process

Introduction to Gaussian Process Introduction to Gaussian Process CS 778 Chris Tensmeyer CS 478 INTRODUCTION 1 What Topic? Machine Learning Regression Bayesian ML Bayesian Regression Bayesian Non-parametric Gaussian Process (GP) GP Regression

More information

Primal-Dual Monotone Kernel Regression

Primal-Dual Monotone Kernel Regression Primal-Dual Monotone Kernel Regression K. Pelckmans, M. Espinoza, J. De Brabanter, J.A.K. Suykens, B. De Moor K.U. Leuven, ESAT-SCD-SISTA Kasteelpark Arenberg B-3 Leuven (Heverlee, Belgium Tel: 32/6/32

More information

Discussion About Nonlinear Time Series Prediction Using Least Squares Support Vector Machine

Discussion About Nonlinear Time Series Prediction Using Least Squares Support Vector Machine Commun. Theor. Phys. (Beijing, China) 43 (2005) pp. 1056 1060 c International Academic Publishers Vol. 43, No. 6, June 15, 2005 Discussion About Nonlinear Time Series Prediction Using Least Squares Support

More information

Neural Networks. Prof. Dr. Rudolf Kruse. Computational Intelligence Group Faculty for Computer Science

Neural Networks. Prof. Dr. Rudolf Kruse. Computational Intelligence Group Faculty for Computer Science Neural Networks Prof. Dr. Rudolf Kruse Computational Intelligence Group Faculty for Computer Science kruse@iws.cs.uni-magdeburg.de Rudolf Kruse Neural Networks 1 Supervised Learning / Support Vector Machines

More information

Learning From Data Lecture 25 The Kernel Trick

Learning From Data Lecture 25 The Kernel Trick Learning From Data Lecture 25 The Kernel Trick Learning with only inner products The Kernel M. Magdon-Ismail CSCI 400/600 recap: Large Margin is Better Controling Overfitting Non-Separable Data 0.08 random

More information

Machine Learning (BSMC-GA 4439) Wenke Liu

Machine Learning (BSMC-GA 4439) Wenke Liu Machine Learning (BSMC-GA 4439) Wenke Liu 02-01-2018 Biomedical data are usually high-dimensional Number of samples (n) is relatively small whereas number of features (p) can be large Sometimes p>>n Problems

More information

ML (cont.): SUPPORT VECTOR MACHINES

ML (cont.): SUPPORT VECTOR MACHINES ML (cont.): SUPPORT VECTOR MACHINES CS540 Bryan R Gibson University of Wisconsin-Madison Slides adapted from those used by Prof. Jerry Zhu, CS540-1 1 / 40 Support Vector Machines (SVMs) The No-Math Version

More information

Support Vector Machines for Classification and Regression. 1 Linearly Separable Data: Hard Margin SVMs

Support Vector Machines for Classification and Regression. 1 Linearly Separable Data: Hard Margin SVMs E0 270 Machine Learning Lecture 5 (Jan 22, 203) Support Vector Machines for Classification and Regression Lecturer: Shivani Agarwal Disclaimer: These notes are a brief summary of the topics covered in

More information

Linear vs Non-linear classifier. CS789: Machine Learning and Neural Network. Introduction

Linear vs Non-linear classifier. CS789: Machine Learning and Neural Network. Introduction Linear vs Non-linear classifier CS789: Machine Learning and Neural Network Support Vector Machine Jakramate Bootkrajang Department of Computer Science Chiang Mai University Linear classifier is in the

More information

Iteratively Reweighted Least Square for Asymmetric L 2 -Loss Support Vector Regression

Iteratively Reweighted Least Square for Asymmetric L 2 -Loss Support Vector Regression Iteratively Reweighted Least Square for Asymmetric L 2 -Loss Support Vector Regression Songfeng Zheng Department of Mathematics Missouri State University Springfield, MO 65897 SongfengZheng@MissouriState.edu

More information

Projection Penalties: Dimension Reduction without Loss

Projection Penalties: Dimension Reduction without Loss Yi Zhang Machine Learning Department, Carnegie Mellon University Jeff Schneider The Robotics Institute, Carnegie Mellon University yizhang1@cs.cmu.edu schneide@cs.cmu.edu Abstract Dimension reduction is

More information

Support Vector Machines for Classification and Regression

Support Vector Machines for Classification and Regression CIS 520: Machine Learning Oct 04, 207 Support Vector Machines for Classification and Regression Lecturer: Shivani Agarwal Disclaimer: These notes are designed to be a supplement to the lecture. They may

More information

Applied inductive learning - Lecture 7

Applied inductive learning - Lecture 7 Applied inductive learning - Lecture 7 Louis Wehenkel & Pierre Geurts Department of Electrical Engineering and Computer Science University of Liège Montefiore - Liège - November 5, 2012 Find slides: http://montefiore.ulg.ac.be/

More information

Classification. The goal: map from input X to a label Y. Y has a discrete set of possible values. We focused on binary Y (values 0 or 1).

Classification. The goal: map from input X to a label Y. Y has a discrete set of possible values. We focused on binary Y (values 0 or 1). Regression and PCA Classification The goal: map from input X to a label Y. Y has a discrete set of possible values We focused on binary Y (values 0 or 1). But we also discussed larger number of classes

More information

Support Vector Machines and Kernel based Learning

Support Vector Machines and Kernel based Learning Support Vector Machines and Kernel based Learning Johan Suykens K.U. Leuven, ESAT-SCD-SISTA Kasteelpark Arenberg 1 B-31 Leuven (Heverlee), Belgium Tel: 32/16/32 18 2 - Fa: 32/16/32 19 7 Email: johan.suykens@esat.kuleuven.ac.be

More information

Stat542 (F11) Statistical Learning. First consider the scenario where the two classes of points are separable.

Stat542 (F11) Statistical Learning. First consider the scenario where the two classes of points are separable. Linear SVM (separable case) First consider the scenario where the two classes of points are separable. It s desirable to have the width (called margin) between the two dashed lines to be large, i.e., have

More information

Support Vector Machines and Kernel based Learning

Support Vector Machines and Kernel based Learning Support Vector Machines and Kernel based Learning Johan Suykens K.U. Leuven, ESAT-SCD-SISTA Kasteelpark Arenberg B-3 Leuven (Heverlee), Belgium Tel: 32/6/32 8 2 - Fa: 32/6/32 9 7 Email: johan.suykens@esat.kuleuven.ac.be

More information

COMS 4721: Machine Learning for Data Science Lecture 10, 2/21/2017

COMS 4721: Machine Learning for Data Science Lecture 10, 2/21/2017 COMS 4721: Machine Learning for Data Science Lecture 10, 2/21/2017 Prof. John Paisley Department of Electrical Engineering & Data Science Institute Columbia University FEATURE EXPANSIONS FEATURE EXPANSIONS

More information

Lecture Notes on Support Vector Machine

Lecture Notes on Support Vector Machine Lecture Notes on Support Vector Machine Feng Li fli@sdu.edu.cn Shandong University, China 1 Hyperplane and Margin In a n-dimensional space, a hyper plane is defined by ω T x + b = 0 (1) where ω R n is

More information

Kernel Methods. Outline

Kernel Methods. Outline Kernel Methods Quang Nguyen University of Pittsburgh CS 3750, Fall 2011 Outline Motivation Examples Kernels Definitions Kernel trick Basic properties Mercer condition Constructing feature space Hilbert

More information

Support Vector Machine for Classification and Regression

Support Vector Machine for Classification and Regression Support Vector Machine for Classification and Regression Ahlame Douzal AMA-LIG, Université Joseph Fourier Master 2R - MOSIG (2013) November 25, 2013 Loss function, Separating Hyperplanes, Canonical Hyperplan

More information

Modifying A Linear Support Vector Machine for Microarray Data Classification

Modifying A Linear Support Vector Machine for Microarray Data Classification Modifying A Linear Support Vector Machine for Microarray Data Classification Prof. Rosemary A. Renaut Dr. Hongbin Guo & Wang Juh Chen Department of Mathematics and Statistics, Arizona State University

More information

Support Vector Machine

Support Vector Machine Andrea Passerini passerini@disi.unitn.it Machine Learning Support vector machines In a nutshell Linear classifiers selecting hyperplane maximizing separation margin between classes (large margin classifiers)

More information

Max Margin-Classifier

Max Margin-Classifier Max Margin-Classifier Oliver Schulte - CMPT 726 Bishop PRML Ch. 7 Outline Maximum Margin Criterion Math Maximizing the Margin Non-Separable Data Kernels and Non-linear Mappings Where does the maximization

More information

Kernel Method: Data Analysis with Positive Definite Kernels

Kernel Method: Data Analysis with Positive Definite Kernels Kernel Method: Data Analysis with Positive Definite Kernels 2. Positive Definite Kernel and Reproducing Kernel Hilbert Space Kenji Fukumizu The Institute of Statistical Mathematics. Graduate University

More information