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

Size: px
Start display at page:

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

Transcription

1 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 Nonsmooth optimization in machine learning, Liege, March (joint work with Marco Signoretto, Quoc Tran Dinh, Lieven De Lathauwer) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens

2 neuroscience: Learning with matrices and tensors EEG data (time samples frequency electrodes) computer vision: image (/video) compression/completion/ (pixel illumination expression ) web mining: analyze users behaviors (users queries webpages) 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 [Signoretto M., Tran Dinh Q., De Lathauwer L., Suykens J.A.K., Learning with Tensors: a Framework Based on Convex Optimization and Spectral Regularization, 2011] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 1

3 Overview Sparsity Matrix completion and tensor completion Learning with matrices and low rank penalty Learning with tensors Optimization algorithms Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 1

4 Learning with 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 Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 2

5 Sparsity in machine learning through loss function: model ŷ = i α ik(x, x i ) + b min w T w + γ i L(e i ) ε 0 +ε sparse α through regularization: model ŷ = w T x + b min j w j + γ i e 2 i sparse w Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 3

6 Underdetermined linear system: Sparsity (1) Ax = b A R n m, n < m Minimum norm solution: Sparsest solution: min x x 2 2 s.t. Ax = b ˆx = A T (AA T ) 1 b (P 0 ) min x x 0 s.t. Ax = b (with x 0 = #{i : x i 0}) Alternatives: l p -norms x p = ( i x i p ) 1/p (P p ) min x x p s.t. Ax = b Nonconvex for 0 < p < 1, convex for p = 1. Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 4

7 Sparsity (2) Mutual coherence: µ(a) = max 1 k,j m, k j a T k a j a k 2 a j 2 For a full rank A R n m,n < m, if a solution x exists satisfying x 0 < 1 2 (1 + 1 µ(a) ) it is both the unique solution of (P 1 ) and (P 0 ). Restricted Isometry Property (RIP): Matrix A R n m has by definition RIP(δ,k) if each submatrix A I (by combining at most k columns of A) has its nonzero singular values bounded between 1 δ and 1 + δ. Matrix A with RIP(0.41; 2k) implies that (P 1 ) and (P 0 ) have identical solutions on all k-sparse vectors. [Bruckstein et al., SIAM Review, 2009; Candes & Tao, 2005; Donoho & Elad, 2003;...] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 5

8 Learning with 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 Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 6

9 Matrix completion: example Given image (80 % missing entries) [experiments by M. Signoretto] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 7

10 Matrix completion: example Given image (80 % missing entries) and completed image [experiments by M. Signoretto] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 7

11 Matrix completion: example Given image (40 % missing entries) [experiments by M. Signoretto] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 7

12 Matrix completion: example Given image (40 % missing entries) and completed image [experiments by M. Signoretto] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 7

13 Matrix completion: example Original image Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 7

14 Matrix completion (1) Given: matrix X with missing entries Goal: complete the missing entries Assumption: assume that X has low rank min X X subject to X ij = Y ij, i, j S given values Y ij with i, j S a subset of all entries of the matrix Nuclear norm X = i σ i with σ i the singular values of X (singular value decomposition: X = i σ iu i v T i ) X is convex envelope of rankx on {X : X 1} [Fazel, 2002] X X F X r X F r X [Recht et al., 2010] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 8

15 Matrix completion (2) This can be written as an SDP problem (semidefinite program): min tr(w 1 ) + tr(w 2 ) X,W 1,W 2 subject to X ij = Y ij i, j S [ ] W1 X X 0 W 2 The nuclear norm plays a similar role as the l 1 norm, at the matrix level. [Fazel et al., 2001; Candes & Recht, 2009] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 9

16 Matrix completion: RIP property Consider: (P 0 ) : min rank X s.t. A(X) = b (P 1 ) : min X s.t. A(X) = b r-restricted isometry constant: smallest number δ r (A) such that 1 δ r (A) A(X) X F 1 + δ r (A) holds for all X of rank at most r, with A : R m n R p a linear map. Suppose that δ 2r < 1 for integer r 1. Then solution to (P 0 ) is the only matrix of rank at most r satisfying A(X) = b. Suppose that r 1, is such that δ 5r < 1 10, then solution to (P 1) equals solution to (P 0 ). [Recht, Fazel, Parrilo, Siam Rev, 2010] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 10

17 Tensor completion Given: N-th order tensor X R I 1.. I N with missing entries Goal: complete the missing entries Assumption: assume that X has low rank with min X X subject to X i1 i 2...i N = Y i1 i 2...i N, i 1,i 2,...,i N S given entries Y i1 i 2...i N with i 1,i 2,...,i N S a subset of the tensor Nuclear norm X = 1 N matrix unfolding n N N X n with X n the n-th mode [Signoretto M., Van De Plas R., De Moor B., Suykens J.A.K., IEEE-SPL, 2011; Gandy et al., 2011, Tomioka et al., 2011] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 11

18 Mass spectral imaging - digital staining Data tensor: pixels 6490 variables/spectrum Given partial labelling (4 classes), SVM prediction on all pixels cerebellar cortex - Ammon s horn section of hippocampus - cauda-putamen - lateral ventricle area [Luts J., Ojeda F., Van de Plas R., De Moor B., Van Huffel S., Suykens J.A.K., ACA 2010] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 12

19 Tensor completion on mass spectral imaging Mass spectral imaging: sagittal section mouse brain [data: E. Waelkens, R. Van de Plas] Tensor completion using nuclear norm regularization [Signoretto et al., IEEE-SPL, 2011] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 13

20 Multichannel EEG for patient-specific seizure detection The electroencephalogram (EEG) measures the electrical activity of the brain and is a well-established technique in epilepsy diagnosis and monitoring. Automatic seizure detection would drastically decrease the workload of clinicians; EEG can provide accurate information about the onset of the seizure. As the seizure spreads quickly through the brain, the early detection of the seizure is essential. [Hunyadi B., Signoretto M., Van Paesschen W., Suykens J., Van Huffel S., De Vos A., Clinical Neurophysiology, 2012] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 14

21 Extracted features: Time domain features: Number zero crossings, max & min 4. Skewness (skew) 5. Kurtosis (kurt) 6. Root mean square amplitude (rmsa) Frequency domain features: 7. Total power (TP) 8. Peak frequency (PF) Feature-channel matrix Mean and normalized power in frequency bands: delta: 13 Hz (D, nd), theta: 48 Hz (T, nt), FP1 F7 F7 T7 T7 P7 P7 O1 FP1 F3 F3 C3 C3 P3 P3 O1 FP2 F4 F4 C4 C4 P4 P4 O2 FP2 F8 F8 T8 T8 P8 P8 O2 FZ CZ CZ PZ P7 T7 T7 FT9 FT9 FT10 FT10 T8 T8 P8 alpha: 913 Hz (A, na), beta: 1420 Hz (B, nb) 365 uv Time (sec) EEG data: CHB-MIT database - scalp EEG recordings, 23 pediatric patients, 18 channels Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 15

22 Model with nuclear norm regularization Synchronization between EEG channels is a generally occurring characteristic. Representing the data in matrix form allows to exploit the common information among the channels. Model: (per patient) ŷ = W,X + b where W,X = ij W ijx ij with X,W R d p, d the number of features, p number of channels. Classifier with decision rule sign[ŷ] Training from given data {(X k, y k )} N k=1 : min W,b N (y k ŷ k ) 2 + µ W k=1 with nuclear norm W = i σ i with singular values σ i ; the labels ±1 correspond to seizure and non-seizure epoch. Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 16

23 Multichannel EEG for patient-specific seizure detection [Hunyadi B., Signoretto M., Van Paesschen W., Suykens J., Van Huffel S., De Vos A., Clinical Neurophysiology, 2012] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 17

24 Learning with 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 Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 18

25 Tensors N-th order tensor A R I 1 I 2 I N inner product: A, B := i 1 i 2 i N A i1 i 2 i N B i1 i 2 i N norm: A := A, A n mode vector: obtained by varying i n and keeping other indices fixed n rank rank n (A): dimension of space spanned by n mode vectors rank-(r 1, r 2,...,r N ) tensor: tensor for which r n = rank n (A) for n N N multilinear rank: N tuple (r 1,r 2,...,r N ) { R N : A = r N R u (1) r u (2) r } : u r (n) R I n r N R, n N N rank: rank(a) := arg min u (N) r property: rank n (A) rank(a) n special case of matrix: rank 1 (A) = rank 2 (A) = rank(a) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 19

26 Tensors N-th order tensor A R I 1 I 2 I N inner product: A, B := i 1 i 2 i N A i1 i 2 i N B i1 i 2 i N norm: A := A, A n mode vector: obtained by varying i n and keeping other indices fixed n rank rank n (A): dimension of space spanned by n mode vectors rank-(r 1, r 2,...,r N ) tensor: tensor for which r n = rank n (A) for n N N multilinear rank: N tuple (r 1,r 2,...,r N ) { R N : A = r N R u (1) r u (2) r } : u r (n) R I n r N R, n N N rank: rank(a) := arg min u (N) r property: rank n (A) rank(a) n special case of matrix: rank 1 (A) = rank 2 (A) = rank(a) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 20

27 Tensors N-th order tensor A R I 1 I 2 I N inner product: A, B := i 1 i 2 i N A i1 i 2 i N B i1 i 2 i N norm: A := A, A n mode vector: obtained by varying i n and keeping other indices fixed n rank rank n (A): dimension of space spanned by n mode vectors rank-(r 1, r 2,...,r N ) tensor: tensor for which r n = rank n (A) for n N N multilinear rank: N tuple (r 1,r 2,...,r N ) { R N : A = r N R u (1) r u (2) r } : u r (n) R I n r N R, n N N rank: rank(a) := arg min u (N) r property: rank n (A) rank(a) n special case of matrix: rank 1 (A) = rank 2 (A) = rank(a) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 21

28 Mode unfoldings of a tensor n mode unfolding A n R I n J (matricization): matrix whose columns are the n mode vectors with J := j N N \{n} I j n mode unfolding n : R I 1 I 2 I N R I n J. refolding: n : R I n J R I 1 I 2 I N property: rank n (A) = rank(a n ) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 22

29 Multilinear SVD (1) [De Lathauwer L., De Moor B., Vandewalle J., 2000] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 23

30 Multilinear SVD (2) n mode product A n U R I 1 I 2 I n 1 J n I n+1 I N : product of tensor A R I 1 I 2 I N by matrix U R J n I n multilinear SVD: with A = S 1 U (1) 2 U (2) 3 N U (N) core tensor S R I 1 I 2 I N U (n) R I n I n a matrix of n mode singular vectors, i.e., left singular vectors of the n mode unfolding W n with SVD A n = U (n) diag(σ(a n ))V (n) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 24

31 Inductive and transductive learning transductive learning with tensors inductive learning with tensors soft-completion data: partially specified input data tensor and matrix of target labels data: pairs of fully specified input features and vectors of target labels output: latent features and missing labels output: models for out-of-sample evaluations of multiple tasks hard-completion data: output: pairs of fully specified input features and vectors of target labels missing input data [Signoretto M., Tran Dinh Q., De Lathauwer L., Suykens J.A.K., 2011] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 25

32 Inductive learning with tensors: setting Training data D N = {( X (n),y (n)) R D 1 D 2 D M R T : n N N } n = 1,...,N training data t = 1,...,T outputs (tasks) M th order input data tensor Model Assumptions: ỹ t = W (t), X + b t, t = 1,...,T X = X + E with X a rank-(r 1,r 2,...,r M ) tensor for core tensors: W (t), X = S W (t), S X low multilinear rank in W (t) = S W (t) 1 U 1 2 U 2 M U M target lables y t generated according to p(y t ỹ t ) = 1/(1 + exp( y t ỹ t )) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 26

33 Inductive learning with tensors: training Penalized empirical risk minimization: min W,b f DN (W, b) + m N M+1 λ m W m with misclassification error e.g. based on logistic loss: f DN : (W,b) n N N t N T log ( ( ))) 1 + exp y (n) t ( X (n), W (t) + b t gives a predictive model, applicable to input data X beyond the training data Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 27

34 Transductive learning with X and Y completion Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 28

35 Transductive learning with tensors: setting Tensors X R D 1 D 2 D M N and Y = [y (1) y (2) y (N) ] R T N Missing entries both in X and Y: S X, S Y : index sets of observed entries of X and Y S SX, S SY : sampling operators related to the index sets Implicit model: ỹ (n) t = W (t), X (n) + b t, t = 1,...,T Assumptions: X = X + E with X a rank-(r 1,r 2,...,r M, r M+1 ) tensor targets y t generated according to p(y tn ỹ tn ) = 1/(1 + exp( y tn ỹ tn )) rank ([ X M+1,Ỹ ]) r M+1 min(n, J + T) with J = j N M D j Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 29

36 Transductive learning with tensors: estimation Estimation of X, Ỹ,b: min ( X,Ỹ,b) V f λ0 ( X,Ỹ, b) + m N M λ m X m + λ M+1 [ X M+1, Ỹ ] objective function: V has module spaces ( R ) D 1 D 2 D M N ( R T N) R T and inner product ( X 1,Ỹ1, b 1 ), ( X 2,Ỹ2, b 2 ) V = X 1, X 2 + Ỹ1,Ỹ2 + b 1, b 2 objective f λ0 ( X,Ỹ, b) = fx y ( X) + λ 0 f (Ỹ, b) with f x : X p N P l x ((Ω S X X) p,zp) x f y : (Ỹ,b) q N Q l y ((Ω SỸ (Ỹ + b 1 J)) q,zq) y losses e.g. l x : (u,v) 1 2 (u v)2, l y : (u,v) log(1 + exp( uv)) z x, z y are vectors of the observed entries Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 30

37 Transductive soft completion: Olivetti faces original input data matrix-sc tensor-sc true label: 5 predicted: 3 predicted: 5 original input data matrix-sc tensor-sc true label: 5 predicted: 3 predicted: 5 original input data matrix-sc tensor-sc true label: 3 predicted: 3 predicted: 3 Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 31

38 Impainting color images by hard completion original given image completed Tensor: mode 1 and 2: pixel space, mode 3: 8-bit RGB color information Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 32

39 Impainting color images by hard completion original given image completed Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 33

40 Impainting color images by hard completion original given image completed Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 33

41 Impainting color images by hard completion original given image completed Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 33

42 Impainting color images by hard completion original given image completed Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 33

43 Optimization algorithm (1) The learning problems are instances of the following convex optimization problem on an abstract vector space: min w W f(w) + ḡ(w) subject to w C with - f: convex and differentiable functional - f is L f-lipschitz: f(w) f(v) W L f w v W w, v W ; - ḡ: convex but possibly non-differentiable functional - C W is a is non-empty, closed and convex set Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 34

44 Optimization algorithm (2) Problem restatement min w W h(w) = f(w) + ḡ(w) + δ C(w), δ C : w { 0, if w C, otherwise Proximity operator x (t+1) = prox τ h (x (t)) with prox τ h : x arg min w W h(w) + 1 2τ w x 2. Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 35

45 Optimization algorithm (3) Operator splitting approach: split h(w) = f(w) + ḡ(w) + δ C(w) into f(w) + δ C(w) and non-smooth term ḡ(w) Douglas-Rachford splitting: y (k) = arg min x C f (x) + 1 2τ r (k) = prox τḡ (2y (k) w (k) ) x w (k) 2 W (solved inexactly) w (k+1) = w (k) + γ (k) ( r (k) y (k)) Projection onto C Proof of convergence for sequence {y (k) } k Stopping criterion based on h Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 36

46 Note: matrix case - Singular Value Thresholding Given matrix Y, finding the solution to min X 1 2 X Y 2 F + λ X with λ > 0, is given by a shrinkage operation on singular values of Y : prox tr λ (Y ) = U max(s λi,0)v T [Cai et al., 2008; Tomioka et al., 2011] Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 37

47 Tensor case Learning problems involve the tensor modes: λ m W m m N M+1 Consider space W with cartesian product W 1 W 2 W I and inner product x,y = i N I x i, y i i. Assume function ḡ : W R defined by ḡ : (x 1, x 2,...,x I ) i N I g i (x i ) where for any i N I, g i : W i R is convex. Then we have: proxḡ(x) = ( prox g1 (x 1 ),prox g2 (x 2 ),,prox gi (x I ) ) Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 38

48 Duplication - transductive learning case C := Duplication of the tensors leads to considering the set: {( X [1], X [2],..., X [M], X [M+1],Ỹ, b) W : X[1] = X [2] =... = X [M+1] } This gives the problem statement: min ( X [1], X [2],..., X [M], X [M+1],Ỹ,b) W subject to f( X [1],..., X [M+1], Ỹ,b) + ḡ( X [1],..., X [M+1],Ỹ ) ( X [1],..., X [M+1], Ỹ,b) C Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 39

49 Prox and tensor modes We apply prox τḡ ( X [1],..., X [M+1], Ỹ ) = (prox τλ1 σ( 1 ) 1 ( X [1] ),, prox τλm σ( M ) 1 ( X [M] ),Z 1, Z 2 ) where [Z 1 ( X, Ỹ ),Z 2( X, Ỹ )] is a partitioning of Z( X,Ỹ ) = Udiag ( prox τλm+1 σ( ) 1 ([ X M+1, Ỹ ])) V with prox λ σ( n ) 1 (W) = (U (n) diag(d λ )V (n) ) n and (d λ ) i := max(σ i (W n ) λ, 0). Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 40

50 Conclusions Sparsity: from vectors to matrices from matrices to tensors Transductive and inductive learning with matrices/tensors: going beyond matrix/tensor completion Further details: Signoretto M., Tran Dinh Q., De Lathauwer L., Suykens J.A.K., Learning with Tensors: a Framework Based on Convex Optimization and Spectral Regularization, 2011 Software: Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 41

51 Acknowledgements Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 42

52 Thank you Learning with matrix and tensor based models using low-rank penalties - Johan Suykens 43

ECE 8201: Low-dimensional Signal Models for High-dimensional Data Analysis

ECE 8201: Low-dimensional Signal Models for High-dimensional Data Analysis ECE 8201: Low-dimensional Signal Models for High-dimensional Data Analysis Lecture 7: Matrix completion Yuejie Chi The Ohio State University Page 1 Reference Guaranteed Minimum-Rank Solutions of Linear

More information

CSC 576: Variants of Sparse Learning

CSC 576: Variants of Sparse Learning CSC 576: Variants of Sparse Learning Ji Liu Department of Computer Science, University of Rochester October 27, 205 Introduction Our previous note basically suggests using l norm to enforce sparsity in

More information

ECE G: Special Topics in Signal Processing: Sparsity, Structure, and Inference

ECE G: Special Topics in Signal Processing: Sparsity, Structure, and Inference ECE 18-898G: Special Topics in Signal Processing: Sparsity, Structure, and Inference Low-rank matrix recovery via convex relaxations Yuejie Chi Department of Electrical and Computer Engineering Spring

More information

Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms

Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms Adrien Todeschini Inria Bordeaux JdS 2014, Rennes Aug. 2014 Joint work with François Caron (Univ. Oxford), Marie

More information

A Characterization of Sampling Patterns for Union of Low-Rank Subspaces Retrieval Problem

A Characterization of Sampling Patterns for Union of Low-Rank Subspaces Retrieval Problem A Characterization of Sampling Patterns for Union of Low-Rank Subspaces Retrieval Problem Morteza Ashraphijuo Columbia University ashraphijuo@ee.columbia.edu Xiaodong Wang Columbia University wangx@ee.columbia.edu

More information

A Fast Augmented Lagrangian Algorithm for Learning Low-Rank Matrices

A Fast Augmented Lagrangian Algorithm for Learning Low-Rank Matrices A Fast Augmented Lagrangian Algorithm for Learning Low-Rank Matrices Ryota Tomioka 1, Taiji Suzuki 1, Masashi Sugiyama 2, Hisashi Kashima 1 1 The University of Tokyo 2 Tokyo Institute of Technology 2010-06-22

More information

Minimizing the Difference of L 1 and L 2 Norms with Applications

Minimizing the Difference of L 1 and L 2 Norms with Applications 1/36 Minimizing the Difference of L 1 and L 2 Norms with Department of Mathematical Sciences University of Texas Dallas May 31, 2017 Partially supported by NSF DMS 1522786 2/36 Outline 1 A nonconvex approach:

More information

A Randomized Approach for Crowdsourcing in the Presence of Multiple Views

A Randomized Approach for Crowdsourcing in the Presence of Multiple Views A Randomized Approach for Crowdsourcing in the Presence of Multiple Views Presenter: Yao Zhou joint work with: Jingrui He - 1 - Roadmap Motivation Proposed framework: M2VW Experimental results Conclusion

More information

Adaptive one-bit matrix completion

Adaptive one-bit matrix completion Adaptive one-bit matrix completion Joseph Salmon Télécom Paristech, Institut Mines-Télécom Joint work with Jean Lafond (Télécom Paristech) Olga Klopp (Crest / MODAL X, Université Paris Ouest) Éric Moulines

More information

Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms

Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms François Caron Department of Statistics, Oxford STATLEARN 2014, Paris April 7, 2014 Joint work with Adrien Todeschini,

More information

Exact Low-rank Matrix Recovery via Nonconvex M p -Minimization

Exact Low-rank Matrix Recovery via Nonconvex M p -Minimization Exact Low-rank Matrix Recovery via Nonconvex M p -Minimization Lingchen Kong and Naihua Xiu Department of Applied Mathematics, Beijing Jiaotong University, Beijing, 100044, People s Republic of China E-mail:

More information

A new truncation strategy for the higher-order singular value decomposition

A new truncation strategy for the higher-order singular value decomposition A new truncation strategy for the higher-order singular value decomposition Nick Vannieuwenhoven K.U.Leuven, Belgium Workshop on Matrix Equations and Tensor Techniques RWTH Aachen, Germany November 21,

More information

Lecture 2 Part 1 Optimization

Lecture 2 Part 1 Optimization Lecture 2 Part 1 Optimization (January 16, 2015) Mu Zhu University of Waterloo Need for Optimization E(y x), P(y x) want to go after them first, model some examples last week then, estimate didn t discuss

More information

Low-Rank Matrix Recovery

Low-Rank Matrix Recovery ELE 538B: Mathematics of High-Dimensional Data Low-Rank Matrix Recovery Yuxin Chen Princeton University, Fall 2018 Outline Motivation Problem setup Nuclear norm minimization RIP and low-rank matrix recovery

More information

Lecture 9: September 28

Lecture 9: September 28 0-725/36-725: Convex Optimization Fall 206 Lecturer: Ryan Tibshirani Lecture 9: September 28 Scribes: Yiming Wu, Ye Yuan, Zhihao Li Note: LaTeX template courtesy of UC Berkeley EECS dept. Disclaimer: These

More information

Some tensor decomposition methods for machine learning

Some tensor decomposition methods for machine learning Some tensor decomposition methods for machine learning Massimiliano Pontil Istituto Italiano di Tecnologia and University College London 16 August 2016 1 / 36 Outline Problem and motivation Tucker decomposition

More information

Recovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies

Recovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies July 12, 212 Recovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies Morteza Mardani Dept. of ECE, University of Minnesota, Minneapolis, MN 55455 Acknowledgments:

More information

Sparse regression. Optimization-Based Data Analysis. Carlos Fernandez-Granda

Sparse regression. Optimization-Based Data Analysis.   Carlos Fernandez-Granda Sparse regression Optimization-Based Data Analysis http://www.cims.nyu.edu/~cfgranda/pages/obda_spring16 Carlos Fernandez-Granda 3/28/2016 Regression Least-squares regression Example: Global warming Logistic

More information

Fast algorithms for informed source separation

Fast algorithms for informed source separation Fast algorithms for informed source separation Augustin Lefèvre augustin.lefevre@uclouvain.be September, 10th 2013 Source separation in 5 minutes Recover source estimates from a mixed signal We consider

More information

Statistical Performance of Convex Tensor Decomposition

Statistical Performance of Convex Tensor Decomposition Slides available: h-p://www.ibis.t.u tokyo.ac.jp/ryotat/tensor12kyoto.pdf Statistical Performance of Convex Tensor Decomposition Ryota Tomioka 2012/01/26 @ Kyoto University Perspectives in Informatics

More information

Robust Principal Component Analysis

Robust Principal Component Analysis ELE 538B: Mathematics of High-Dimensional Data Robust Principal Component Analysis Yuxin Chen Princeton University, Fall 2018 Disentangling sparse and low-rank matrices Suppose we are given a matrix M

More information

Lecture: Matrix Completion

Lecture: Matrix Completion 1/56 Lecture: Matrix Completion http://bicmr.pku.edu.cn/~wenzw/bigdata2017.html Acknowledgement: this slides is based on Prof. Jure Leskovec and Prof. Emmanuel Candes s lecture notes Recommendation systems

More information

Learning with primal and dual model representations

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

More information

Optimization methods

Optimization methods Lecture notes 3 February 8, 016 1 Introduction Optimization methods In these notes we provide an overview of a selection of optimization methods. We focus on methods which rely on first-order information,

More information

Proximal Gradient Descent and Acceleration. Ryan Tibshirani Convex Optimization /36-725

Proximal Gradient Descent and Acceleration. Ryan Tibshirani Convex Optimization /36-725 Proximal Gradient Descent and Acceleration Ryan Tibshirani Convex Optimization 10-725/36-725 Last time: subgradient method Consider the problem min f(x) with f convex, and dom(f) = R n. Subgradient method:

More information

Optimization methods

Optimization methods Optimization methods Optimization-Based Data Analysis http://www.cims.nyu.edu/~cfgranda/pages/obda_spring16 Carlos Fernandez-Granda /8/016 Introduction Aim: Overview of optimization methods that Tend to

More information

A direct formulation for sparse PCA using semidefinite programming

A direct formulation for sparse PCA using semidefinite programming A direct formulation for sparse PCA using semidefinite programming A. d Aspremont, L. El Ghaoui, M. Jordan, G. Lanckriet ORFE, Princeton University & EECS, U.C. Berkeley A. d Aspremont, INFORMS, Denver,

More information

CHAPTER 11. A Revision. 1. The Computers and Numbers therein

CHAPTER 11. A Revision. 1. The Computers and Numbers therein CHAPTER A Revision. The Computers and Numbers therein Traditional computer science begins with a finite alphabet. By stringing elements of the alphabet one after another, one obtains strings. A set of

More information

ARock: an algorithmic framework for asynchronous parallel coordinate updates

ARock: an algorithmic framework for asynchronous parallel coordinate updates ARock: an algorithmic framework for asynchronous parallel coordinate updates Zhimin Peng, Yangyang Xu, Ming Yan, Wotao Yin ( UCLA Math, U.Waterloo DCO) UCLA CAM Report 15-37 ShanghaiTech SSDS 15 June 25,

More information

Approximate Low-Rank Tensor Learning

Approximate Low-Rank Tensor Learning Approximate Low-Rank Tensor Learning Yaoliang Yu Dept of Machine Learning Carnegie Mellon University yaoliang@cs.cmu.edu Hao Cheng Dept of Electric Engineering University of Washington kelvinwsch@gmail.com

More information

Compressed Sensing and Robust Recovery of Low Rank Matrices

Compressed Sensing and Robust Recovery of Low Rank Matrices Compressed Sensing and Robust Recovery of Low Rank Matrices M. Fazel, E. Candès, B. Recht, P. Parrilo Electrical Engineering, University of Washington Applied and Computational Mathematics Dept., Caltech

More information

Is the test error unbiased for these programs? 2017 Kevin Jamieson

Is the test error unbiased for these programs? 2017 Kevin Jamieson Is the test error unbiased for these programs? 2017 Kevin Jamieson 1 Is the test error unbiased for this program? 2017 Kevin Jamieson 2 Simple Variable Selection LASSO: Sparse Regression Machine Learning

More information

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images!

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images! Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images! Alfredo Nava-Tudela John J. Benedetto, advisor 1 Happy birthday Lucía! 2 Outline - Problem: Find sparse solutions

More information

SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels

SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels SVMs: Non-Separable Data, Convex Surrogate Loss, Multi-Class Classification, Kernels Karl Stratos June 21, 2018 1 / 33 Tangent: Some Loose Ends in Logistic Regression Polynomial feature expansion in logistic

More information

The convex algebraic geometry of rank minimization

The convex algebraic geometry of rank minimization The convex algebraic geometry of rank minimization Pablo A. Parrilo Laboratory for Information and Decision Systems Massachusetts Institute of Technology International Symposium on Mathematical Programming

More information

Introduction to Compressed Sensing

Introduction to Compressed Sensing Introduction to Compressed Sensing Alejandro Parada, Gonzalo Arce University of Delaware August 25, 2016 Motivation: Classical Sampling 1 Motivation: Classical Sampling Issues Some applications Radar Spectral

More information

New Coherence and RIP Analysis for Weak. Orthogonal Matching Pursuit

New Coherence and RIP Analysis for Weak. Orthogonal Matching Pursuit New Coherence and RIP Analysis for Wea 1 Orthogonal Matching Pursuit Mingrui Yang, Member, IEEE, and Fran de Hoog arxiv:1405.3354v1 [cs.it] 14 May 2014 Abstract In this paper we define a new coherence

More information

Is the test error unbiased for these programs?

Is the test error unbiased for these programs? Is the test error unbiased for these programs? Xtrain avg N o Preprocessing by de meaning using whole TEST set 2017 Kevin Jamieson 1 Is the test error unbiased for this program? e Stott see non for f x

More information

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Alfredo Nava-Tudela ant@umd.edu John J. Benedetto Department of Mathematics jjb@umd.edu Abstract In this project we are

More information

Statistical Data Mining and Machine Learning Hilary Term 2016

Statistical Data Mining and Machine Learning Hilary Term 2016 Statistical Data Mining and Machine Learning Hilary Term 2016 Dino Sejdinovic Department of Statistics Oxford Slides and other materials available at: http://www.stats.ox.ac.uk/~sejdinov/sdmml Naïve Bayes

More information

Compressive Sensing of Sparse Tensor

Compressive Sensing of Sparse Tensor Shmuel Friedland Univ. Illinois at Chicago Matheon Workshop CSA2013 December 12, 2013 joint work with Q. Li and D. Schonfeld, UIC Abstract Conventional Compressive sensing (CS) theory relies on data representation

More information

Going off the grid. Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison

Going off the grid. Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison Going off the grid Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison Joint work with Badri Bhaskar Parikshit Shah Gonnguo Tang We live in a continuous world... But we work

More information

An Extended Frank-Wolfe Method, with Application to Low-Rank Matrix Completion

An Extended Frank-Wolfe Method, with Application to Low-Rank Matrix Completion An Extended Frank-Wolfe Method, with Application to Low-Rank Matrix Completion Robert M. Freund, MIT joint with Paul Grigas (UC Berkeley) and Rahul Mazumder (MIT) CDC, December 2016 1 Outline of Topics

More information

Master 2 MathBigData. 3 novembre CMAP - Ecole Polytechnique

Master 2 MathBigData. 3 novembre CMAP - Ecole Polytechnique Master 2 MathBigData S. Gaïffas 1 3 novembre 2014 1 CMAP - Ecole Polytechnique 1 Supervised learning recap Introduction Loss functions, linearity 2 Penalization Introduction Ridge Sparsity Lasso 3 Some

More information

Sparse and Low-Rank Matrix Decompositions

Sparse and Low-Rank Matrix Decompositions Forty-Seventh Annual Allerton Conference Allerton House, UIUC, Illinois, USA September 30 - October 2, 2009 Sparse and Low-Rank Matrix Decompositions Venkat Chandrasekaran, Sujay Sanghavi, Pablo A. Parrilo,

More information

Compressed Sensing: a Subgradient Descent Method for Missing Data Problems

Compressed Sensing: a Subgradient Descent Method for Missing Data Problems Compressed Sensing: a Subgradient Descent Method for Missing Data Problems ANZIAM, Jan 30 Feb 3, 2011 Jonathan M. Borwein Jointly with D. Russell Luke, University of Goettingen FRSC FAAAS FBAS FAA Director,

More information

Lecture Notes 10: Matrix Factorization

Lecture Notes 10: Matrix Factorization Optimization-based data analysis Fall 207 Lecture Notes 0: Matrix Factorization Low-rank models. Rank- model Consider the problem of modeling a quantity y[i, j] that depends on two indices i and j. To

More information

Sparse PCA with applications in finance

Sparse PCA with applications in finance Sparse PCA with applications in finance A. d Aspremont, L. El Ghaoui, M. Jordan, G. Lanckriet ORFE, Princeton University & EECS, U.C. Berkeley Available online at www.princeton.edu/~aspremon 1 Introduction

More information

Properties of Matrices and Operations on Matrices

Properties of Matrices and Operations on Matrices Properties of Matrices and Operations on Matrices A common data structure for statistical analysis is a rectangular array or matris. Rows represent individual observational units, or just observations,

More information

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 9. Alternating Direction Method of Multipliers

Shiqian Ma, MAT-258A: Numerical Optimization 1. Chapter 9. Alternating Direction Method of Multipliers Shiqian Ma, MAT-258A: Numerical Optimization 1 Chapter 9 Alternating Direction Method of Multipliers Shiqian Ma, MAT-258A: Numerical Optimization 2 Separable convex optimization a special case is min f(x)

More information

Homework 4. Convex Optimization /36-725

Homework 4. Convex Optimization /36-725 Homework 4 Convex Optimization 10-725/36-725 Due Friday November 4 at 5:30pm submitted to Christoph Dann in Gates 8013 (Remember to a submit separate writeup for each problem, with your name at the top)

More information

A direct formulation for sparse PCA using semidefinite programming

A direct formulation for sparse PCA using semidefinite programming A direct formulation for sparse PCA using semidefinite programming A. d Aspremont, L. El Ghaoui, M. Jordan, G. Lanckriet ORFE, Princeton University & EECS, U.C. Berkeley Available online at www.princeton.edu/~aspremon

More information

Matrix Completion: Fundamental Limits and Efficient Algorithms

Matrix Completion: Fundamental Limits and Efficient Algorithms Matrix Completion: Fundamental Limits and Efficient Algorithms Sewoong Oh PhD Defense Stanford University July 23, 2010 1 / 33 Matrix completion Find the missing entries in a huge data matrix 2 / 33 Example

More information

Oslo Class 6 Sparsity based regularization

Oslo Class 6 Sparsity based regularization RegML2017@SIMULA Oslo Class 6 Sparsity based regularization Lorenzo Rosasco UNIGE-MIT-IIT May 4, 2017 Learning from data Possible only under assumptions regularization min Ê(w) + λr(w) w Smoothness Sparsity

More information

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation Alfredo Nava-Tudela John J. Benedetto, advisor 5/10/11 AMSC 663/664 1 Problem Let A be an n

More information

Conditional Gradient Algorithms for Rank-One Matrix Approximations with a Sparsity Constraint

Conditional Gradient Algorithms for Rank-One Matrix Approximations with a Sparsity Constraint Conditional Gradient Algorithms for Rank-One Matrix Approximations with a Sparsity Constraint Marc Teboulle School of Mathematical Sciences Tel Aviv University Joint work with Ronny Luss Optimization and

More information

Recent Developments in Compressed Sensing

Recent Developments in Compressed Sensing Recent Developments in Compressed Sensing M. Vidyasagar Distinguished Professor, IIT Hyderabad m.vidyasagar@iith.ac.in, www.iith.ac.in/ m vidyasagar/ ISL Seminar, Stanford University, 19 April 2018 Outline

More information

Generalized Higher-Order Tensor Decomposition via Parallel ADMM

Generalized Higher-Order Tensor Decomposition via Parallel ADMM Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence Generalized Higher-Order Tensor Decomposition via Parallel ADMM Fanhua Shang 1, Yuanyuan Liu 2, James Cheng 1 1 Department of

More information

Reweighted Nuclear Norm Minimization with Application to System Identification

Reweighted Nuclear Norm Minimization with Application to System Identification Reweighted Nuclear Norm Minimization with Application to System Identification Karthi Mohan and Maryam Fazel Abstract The matrix ran minimization problem consists of finding a matrix of minimum ran that

More information

Optimization for Learning and Big Data

Optimization for Learning and Big Data Optimization for Learning and Big Data Donald Goldfarb Department of IEOR Columbia University Department of Mathematics Distinguished Lecture Series May 17-19, 2016. Lecture 1. First-Order Methods for

More information

10-725/36-725: Convex Optimization Prerequisite Topics

10-725/36-725: Convex Optimization Prerequisite Topics 10-725/36-725: Convex Optimization Prerequisite Topics February 3, 2015 This is meant to be a brief, informal refresher of some topics that will form building blocks in this course. The content of the

More information

Inverse problems and sparse models (1/2) Rémi Gribonval INRIA Rennes - Bretagne Atlantique, France

Inverse problems and sparse models (1/2) Rémi Gribonval INRIA Rennes - Bretagne Atlantique, France Inverse problems and sparse models (1/2) Rémi Gribonval INRIA Rennes - Bretagne Atlantique, France remi.gribonval@inria.fr Structure of the tutorial Session 1: Introduction to inverse problems & sparse

More information

MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications. Class 08: Sparsity Based Regularization. Lorenzo Rosasco

MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications. Class 08: Sparsity Based Regularization. Lorenzo Rosasco MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications Class 08: Sparsity Based Regularization Lorenzo Rosasco Learning algorithms so far ERM + explicit l 2 penalty 1 min w R d n n l(y

More information

Data Mining and Matrices

Data Mining and Matrices Data Mining and Matrices 05 Semi-Discrete Decomposition Rainer Gemulla, Pauli Miettinen May 16, 2013 Outline 1 Hunting the Bump 2 Semi-Discrete Decomposition 3 The Algorithm 4 Applications SDD alone SVD

More information

Coordinate Update Algorithm Short Course Operator Splitting

Coordinate Update Algorithm Short Course Operator Splitting Coordinate Update Algorithm Short Course Operator Splitting Instructor: Wotao Yin (UCLA Math) Summer 2016 1 / 25 Operator splitting pipeline 1. Formulate a problem as 0 A(x) + B(x) with monotone operators

More information

σ 11 σ 22 σ pp 0 with p = min(n, m) The σ ii s are the singular values. Notation change σ ii A 1 σ 2

σ 11 σ 22 σ pp 0 with p = min(n, m) The σ ii s are the singular values. Notation change σ ii A 1 σ 2 HE SINGULAR VALUE DECOMPOSIION he SVD existence - properties. Pseudo-inverses and the SVD Use of SVD for least-squares problems Applications of the SVD he Singular Value Decomposition (SVD) heorem For

More information

Stochastic dynamical modeling:

Stochastic dynamical modeling: Stochastic dynamical modeling: Structured matrix completion of partially available statistics Armin Zare www-bcf.usc.edu/ arminzar Joint work with: Yongxin Chen Mihailo R. Jovanovic Tryphon T. Georgiou

More information

A randomized block sampling approach to the canonical polyadic decomposition of large-scale tensors

A randomized block sampling approach to the canonical polyadic decomposition of large-scale tensors A randomized block sampling approach to the canonical polyadic decomposition of large-scale tensors Nico Vervliet Joint work with Lieven De Lathauwer SIAM AN17, July 13, 2017 2 Classification of hazardous

More information

Optimisation Combinatoire et Convexe.

Optimisation Combinatoire et Convexe. Optimisation Combinatoire et Convexe. Low complexity models, l 1 penalties. A. d Aspremont. M1 ENS. 1/36 Today Sparsity, low complexity models. l 1 -recovery results: three approaches. Extensions: matrix

More information

Matrices and Vectors. Definition of Matrix. An MxN matrix A is a two-dimensional array of numbers A =

Matrices and Vectors. Definition of Matrix. An MxN matrix A is a two-dimensional array of numbers A = 30 MATHEMATICS REVIEW G A.1.1 Matrices and Vectors Definition of Matrix. An MxN matrix A is a two-dimensional array of numbers A = a 11 a 12... a 1N a 21 a 22... a 2N...... a M1 a M2... a MN A matrix can

More information

Machine Learning and Computational Statistics, Spring 2017 Homework 2: Lasso Regression

Machine Learning and Computational Statistics, Spring 2017 Homework 2: Lasso Regression Machine Learning and Computational Statistics, Spring 2017 Homework 2: Lasso Regression Due: Monday, February 13, 2017, at 10pm (Submit via Gradescope) Instructions: Your answers to the questions below,

More information

Structured matrix factorizations. Example: Eigenfaces

Structured matrix factorizations. Example: Eigenfaces Structured matrix factorizations Example: Eigenfaces An extremely large variety of interesting and important problems in machine learning can be formulated as: Given a matrix, find a matrix and a matrix

More information

New Ranks for Even-Order Tensors and Their Applications in Low-Rank Tensor Optimization

New Ranks for Even-Order Tensors and Their Applications in Low-Rank Tensor Optimization New Ranks for Even-Order Tensors and Their Applications in Low-Rank Tensor Optimization Bo JIANG Shiqian MA Shuzhong ZHANG January 12, 2015 Abstract In this paper, we propose three new tensor decompositions

More information

This can be 2 lectures! still need: Examples: non-convex problems applications for matrix factorization

This can be 2 lectures! still need: Examples: non-convex problems applications for matrix factorization This can be 2 lectures! still need: Examples: non-convex problems applications for matrix factorization x = prox_f(x)+prox_{f^*}(x) use to get prox of norms! PROXIMAL METHODS WHY PROXIMAL METHODS Smooth

More information

Conditions for Robust Principal Component Analysis

Conditions for Robust Principal Component Analysis Rose-Hulman Undergraduate Mathematics Journal Volume 12 Issue 2 Article 9 Conditions for Robust Principal Component Analysis Michael Hornstein Stanford University, mdhornstein@gmail.com Follow this and

More information

Least Squares Classification

Least Squares Classification Least Squares Classification Stephen Boyd EE103 Stanford University November 4, 2017 Outline Classification Least squares classification Multi-class classifiers Classification 2 Classification data fitting

More information

Rank Determination for Low-Rank Data Completion

Rank Determination for Low-Rank Data Completion Journal of Machine Learning Research 18 017) 1-9 Submitted 7/17; Revised 8/17; Published 9/17 Rank Determination for Low-Rank Data Completion Morteza Ashraphijuo Columbia University New York, NY 1007,

More information

LOW MULTILINEAR RANK TENSOR APPROXIMATION VIA SEMIDEFINITE PROGRAMMING

LOW MULTILINEAR RANK TENSOR APPROXIMATION VIA SEMIDEFINITE PROGRAMMING 17th European Signal Processing Conference (EUSIPCO 2009) Glasgow, Scotland, August 24-28, 2009 LOW MULTILINEAR RANK TENSOR APPROXIMATION VIA SEMIDEFINITE PROGRAMMING Carmeliza Navasca and Lieven De Lathauwer

More information

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2013 PROBLEM SET 2

STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2013 PROBLEM SET 2 STAT 309: MATHEMATICAL COMPUTATIONS I FALL 2013 PROBLEM SET 2 1. You are not allowed to use the svd for this problem, i.e. no arguments should depend on the svd of A or A. Let W be a subspace of C n. The

More information

Convex Optimization Algorithms for Machine Learning in 10 Slides

Convex Optimization Algorithms for Machine Learning in 10 Slides Convex Optimization Algorithms for Machine Learning in 10 Slides Presenter: Jul. 15. 2015 Outline 1 Quadratic Problem Linear System 2 Smooth Problem Newton-CG 3 Composite Problem Proximal-Newton-CD 4 Non-smooth,

More information

Rank minimization via the γ 2 norm

Rank minimization via the γ 2 norm Rank minimization via the γ 2 norm Troy Lee Columbia University Adi Shraibman Weizmann Institute Rank Minimization Problem Consider the following problem min X rank(x) A i, X b i for i = 1,..., k Arises

More information

Robust Principal Component Pursuit via Alternating Minimization Scheme on Matrix Manifolds

Robust Principal Component Pursuit via Alternating Minimization Scheme on Matrix Manifolds Robust Principal Component Pursuit via Alternating Minimization Scheme on Matrix Manifolds Tao Wu Institute for Mathematics and Scientific Computing Karl-Franzens-University of Graz joint work with Prof.

More information

Sparsity in Underdetermined Systems

Sparsity in Underdetermined Systems Sparsity in Underdetermined Systems Department of Statistics Stanford University August 19, 2005 Classical Linear Regression Problem X n y p n 1 > Given predictors and response, y Xβ ε = + ε N( 0, σ 2

More information

Big Data Analytics: Optimization and Randomization

Big Data Analytics: Optimization and Randomization Big Data Analytics: Optimization and Randomization Tianbao Yang Tutorial@ACML 2015 Hong Kong Department of Computer Science, The University of Iowa, IA, USA Nov. 20, 2015 Yang Tutorial for ACML 15 Nov.

More information

Homework 5. Convex Optimization /36-725

Homework 5. Convex Optimization /36-725 Homework 5 Convex Optimization 10-725/36-725 Due Tuesday November 22 at 5:30pm submitted to Christoph Dann in Gates 8013 (Remember to a submit separate writeup for each problem, with your name at the top)

More information

From Compressed Sensing to Matrix Completion and Beyond. Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison

From Compressed Sensing to Matrix Completion and Beyond. Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison From Compressed Sensing to Matrix Completion and Beyond Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison Netflix Prize One million big ones! Given 100 million ratings on a

More information

Lecture 1: September 25

Lecture 1: September 25 0-725: Optimization Fall 202 Lecture : September 25 Lecturer: Geoff Gordon/Ryan Tibshirani Scribes: Subhodeep Moitra Note: LaTeX template courtesy of UC Berkeley EECS dept. Disclaimer: These notes have

More information

An accelerated proximal gradient algorithm for nuclear norm regularized linear least squares problems

An accelerated proximal gradient algorithm for nuclear norm regularized linear least squares problems An accelerated proximal gradient algorithm for nuclear norm regularized linear least squares problems Kim-Chuan Toh Sangwoon Yun March 27, 2009; Revised, Nov 11, 2009 Abstract The affine rank minimization

More information

Variational Functions of Gram Matrices: Convex Analysis and Applications

Variational Functions of Gram Matrices: Convex Analysis and Applications Variational Functions of Gram Matrices: Convex Analysis and Applications Maryam Fazel University of Washington Joint work with: Amin Jalali (UW), Lin Xiao (Microsoft Research) IMA Workshop on Resource

More information

ENGG5781 Matrix Analysis and Computations Lecture 10: Non-Negative Matrix Factorization and Tensor Decomposition

ENGG5781 Matrix Analysis and Computations Lecture 10: Non-Negative Matrix Factorization and Tensor Decomposition ENGG5781 Matrix Analysis and Computations Lecture 10: Non-Negative Matrix Factorization and Tensor Decomposition Wing-Kin (Ken) Ma 2017 2018 Term 2 Department of Electronic Engineering The Chinese University

More information

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation

Course Notes for EE227C (Spring 2018): Convex Optimization and Approximation Course Notes for EE7C (Spring 08): Convex Optimization and Approximation Instructor: Moritz Hardt Email: hardt+ee7c@berkeley.edu Graduate Instructor: Max Simchowitz Email: msimchow+ee7c@berkeley.edu October

More information

Maximum Margin Matrix Factorization

Maximum Margin Matrix Factorization Maximum Margin Matrix Factorization Nati Srebro Toyota Technological Institute Chicago Joint work with Noga Alon Tel-Aviv Yonatan Amit Hebrew U Alex d Aspremont Princeton Michael Fink Hebrew U Tommi Jaakkola

More information

Introduction to Sparsity. Xudong Cao, Jake Dreamtree & Jerry 04/05/2012

Introduction to Sparsity. Xudong Cao, Jake Dreamtree & Jerry 04/05/2012 Introduction to Sparsity Xudong Cao, Jake Dreamtree & Jerry 04/05/2012 Outline Understanding Sparsity Total variation Compressed sensing(definition) Exact recovery with sparse prior(l 0 ) l 1 relaxation

More information

Proximal Methods for Optimization with Spasity-inducing Norms

Proximal Methods for Optimization with Spasity-inducing Norms Proximal Methods for Optimization with Spasity-inducing Norms Group Learning Presentation Xiaowei Zhou Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology

More information

Multi-stage convex relaxation approach for low-rank structured PSD matrix recovery

Multi-stage convex relaxation approach for low-rank structured PSD matrix recovery Multi-stage convex relaxation approach for low-rank structured PSD matrix recovery Department of Mathematics & Risk Management Institute National University of Singapore (Based on a joint work with Shujun

More information

Sparse Covariance Selection using Semidefinite Programming

Sparse Covariance Selection using Semidefinite Programming Sparse Covariance Selection using Semidefinite Programming A. d Aspremont ORFE, Princeton University Joint work with O. Banerjee, L. El Ghaoui & G. Natsoulis, U.C. Berkeley & Iconix Pharmaceuticals Support

More information

Convex Methods for Transduction

Convex Methods for Transduction Convex Methods for Transduction Tijl De Bie ESAT-SCD/SISTA, K.U.Leuven Kasteelpark Arenberg 10 3001 Leuven, Belgium tijl.debie@esat.kuleuven.ac.be Nello Cristianini Department of Statistics, U.C.Davis

More information

ACCELERATED LINEARIZED BREGMAN METHOD. June 21, Introduction. In this paper, we are interested in the following optimization problem.

ACCELERATED LINEARIZED BREGMAN METHOD. June 21, Introduction. In this paper, we are interested in the following optimization problem. ACCELERATED LINEARIZED BREGMAN METHOD BO HUANG, SHIQIAN MA, AND DONALD GOLDFARB June 21, 2011 Abstract. In this paper, we propose and analyze an accelerated linearized Bregman (A) method for solving the

More information

Bindel, Fall 2009 Matrix Computations (CS 6210) Week 8: Friday, Oct 17

Bindel, Fall 2009 Matrix Computations (CS 6210) Week 8: Friday, Oct 17 Logistics Week 8: Friday, Oct 17 1. HW 3 errata: in Problem 1, I meant to say p i < i, not that p i is strictly ascending my apologies. You would want p i > i if you were simply forming the matrices and

More information

Principal Component Analysis

Principal Component Analysis Machine Learning Michaelmas 2017 James Worrell Principal Component Analysis 1 Introduction 1.1 Goals of PCA Principal components analysis (PCA) is a dimensionality reduction technique that can be used

More information