Site menu:
ML & Stats;
Optimization
Sci. Comp.
Mathematik
In the pipeline
- "On some questions of Bellman and Zhang on Hua matrices"
by S. Sra
Status: ■■■■ (Jul 2014); Category:
- "Riemannian dictionary learning and sparse coding"
by A. Cherian, S. Sra
Status: ■ (Jun 2014); Category:
- "Diagonalization of an anti-triangular Cesaró matrix"
by S. Sra
Status: ■■■■■ (Feb 2014); Category:
- "Minimax Optimal Horizon--dependent Priors in Online Binary Prediction with Bernoulli Experts"
by A. Malek, S. Sra, F. Hedyati, P. Bartlett
Status: ■■■ (Jan 2014); Category:
- "Kernels"
by S. Sra and B. Schölkopf
Status: ■ (Dec 2013); Category:
- "Kernels and distances via symmetric polynomials"
by S. Sra
Status: ■■■ (Jun 2013); Category:
- "Improved projection and proximity subroutines for Lp-norms"
by A. Barbero, S. Sra
Status: ■■■; Caterogy:
- "Faster optimization on the manifold of positive definite matrices for the matrix geometric mean"
by S. Sra
Status: ■■■ (Jan 2013); Category
Submitted Articles
- "Conic geometric optimisation on the manifold of positive definite matrices"
by S. Sra, R. Hosseini
arXiv: [1312.1039; math.FA]. Dec 2013
- "Title suppressed for now"
by S. Sra, R. Hosseini
Journal submission (Jun 2014)
- "Title suppressed for now"
by S. Sra, R. Hosseini
Conference submission (Jun 2014)
- "Positive Definite Matrices and the S-Divergence"--Updated version!
by S. Sra
(version of Oct 2012 was originally accepted by SIMAX);
(Dec. 2013) arXiv: [math.FA-1110.1773v4]
- "Modular proximal optimization with application to total variation regularization"
by Á. J. Barbero, S. Sra
Journal article (submitted) (Nov. 2013);
2014
- Riemannian sparse coding for positive definite matrices
by A. Cherian, S. Sra
European Conference on Computer Vision (ECCV) (Jun 2014)
- Fast Newton methods for the group fused lasso.
by M. Wytock, S. Sra, Z. Kolter,
Uncertainty in Artificial Intelligence (UAI) (May 2014)
- "Efficient nearest neighbors via robust sparse hashing"
by A. Cherian, S. Sra, V. Morellas, and N. Papanikolopoulos
IEEE Transactions on Image Processing accepted (Apr 2014);
Preprint: [.pdf];
- "Randomized Nonlinear Component Analysis"
by D. Lopez-Paz, S. Sra, A. Smola, Z. Ghahramani, and B. Schölkopf
International Conf. on Machine Learning (ICML'14); 2014;
![Machine Learning, Data Mining, Statistics](images/ml.png)
arXiv: [1402.0119]
- "Towards stochastic alternating direction method of multipliers"
by S. Azadi and S. Sra
International Conf. on Machine Learning (ICML'14); 2014
Paper: [.pdf];
![Optimization](images/opt.png)
- "Nonconvex proximal splitting: batch and incremental algorithms"
by S. Sra
Invited book chapter in:
Regularization, Optimization, Kernels, and Support Vector Machines
(Editors: J. A.K. Suykens, M. Signoretto, A. Argyriou. (Mar 2014);
MPI-IS-Tech Report #2: [.pdf]
arXiv: [math.OC-1109.0258]
2013
- "Tractable large-scale optimization in machine learning"
by S. Sra
Invited book chapter in:
Tractability Practical Approaches to Hard Problems
(Editors: L. Bordeaux, Y. Hamadi, P. Kohli); Aug 2013.
Preprint: [.pdf (Apr 2012)] (TBD)
- Geometric optimisation on positive definite matrices with application to elliptically contoured distributions
by S. Sra and R. Hosseini
Advances in Neural Information Processing Systems (NIPS) Sep. 2013
Preprint: [.pdf]; Paper: [.pdf];
![Optimization](images/opt.png)
- Reflection methods for user-friendly submodular optimization
by S. Jegelka, F. Bach, and S. Sra
Advances in Neural Information Processing Systems (NIPS) Sep. 2013
Preprint: [.pdf]; Paper: [.pdf];
![Optimization](images/opt.png)
- "Correlation matrix nearness and completion under observation uncertainty"
by C. M. Alaiz, F. Dinuzzo, and S. Sra
IMA Journal of Numerical Analysis (Oct 2013);
Paper: [.pdf];
2012
- "Jensen-Bregman LogDet Divergence for Efficient Similarity Computations on Positive Definite Tensors"
by A. Cherian, S. Sra, A. Banerjee, and N. Papanikolopoulos
IEEE Tran. Pattern. Analy. Mach. Intell. (TPAMI) Dec. 2012;
TR: [.pdf]
- "A new metric on the manifold of kernel matrices with application to matrix geometric means"
by S. Sra
Advances of Neural Information Processing Systems (NIPS) 2012;
[.pdf]
- "Scalable nonconvex inexact proximal splitting"
by S. Sra
  Advances of Neural Information Processing Systems (NIPS) 2012;
[.pdf]
- "The multivariate Watson distribution: Maximum-likelihood estimation and other aspects"
by S. Sra and D. B. Karp
to appear in Journal of Multivariate Analysis (submitted Apr. 2011; accepted Aug 2012);
Preprint: [stat.CO-1104.4422]; official version: [.pdf]
- "Explicit eigenvalues of certain scaled trigonometric matrices"
by S. Sra
Linear Algebra and its Applications;
(Submitted Jan., 2012, accepted Jul 2012)
[.pdf]; preprint: [math.NA-1201.4651]
- "Fast projection onto mixed-norm balls with applications"
by S. Sra
to appear in Data Minining and Knowledge Discovery (DMKD) 2012;
preprint: [stat.ML-1204.1437]
2011
- "A non-monotonic method for large-scale non-negative least squares"
by D. Kim, S. Sra, I. S. Dhillon
(submitted: Nov. 2010; revised: May. 2011; accepted: Dec. 2011)
in Optimization Methods and Software;
Paper: [.pdf]
- "Efficient Similarity Search for Covariance Matrices via the Jensen-Bregman LogDet Divergence"
by A. Cherian, S. Sra, A. Banerjee, and N. Papanikolopoulos
in International Conference on Computer Vision (ICCV) (2011);
Paper: [.pdf]; Bugfix version: [.pdf];
- "Generalized Dictionary Learning for Symmetric Positive Definite Matrices with Application to Nearest Neighbor Retrieval"
by S. Sra, A. Cherian
in European Conf. on Machine Learning (ECML) (2011);
Paper: [.pdf]
- "Fast projections onto L1,q-norm balls for grouped feature selection"
by S. Sra
European Conf. on Machine Learning (ECML) (2011);
(Best Paper Runner up Award)
Paper: [.pdf]
- "Fast Newton-type Methods for Total-Variation with Applications"
by Á. J. Barbero, S. Sra
International Conference on Machine Learning (ICML) June, 2011;
Paper: [.pdf]
- "A short note on parameter approximation for von Mises-Fisher distributions: and a fast implementation of $I_s(x)$"
by S. Sra
(revision of Apr. 2009) Computational Statistics (2011);
Preprint: [.pdf]
- "Optimization for Machine Learning"
by S. Sra, S. Nowozin, S. J. Wright
MIT Press (2011);
(Book at machine learning and optimization users, students, researchers.)
Buy it at: MIT Press or Amazon or Barnes and Noble
- "Projected Newton-type methods in machine learning."
by M. Schmidt, D. Kim, S. Sra
Chapter in: Optimization for Machine Learning. MIT Press (2011);
Preprint: [.pdf]
- "Online Multi-frame Blind Deconvolution with Super-resolution and Saturation Correction"
by M. Hirsch, S. Harmeling, S. Sra, B. Schölkopf
in Astronomy & Astrophysics Feb. (2011);
Paper: [.pdf]
- "Denoising sparse noise via online dictionary learning"
by A. Cherian, S. Sra, N. Papanikolopoulos
in IEEE Conf. Speech Acoustics and Signal Processing (ICASSP), May 2011;
Paper: [.pdf]
2010
- "Sparse inverse covariance estimation using an adaptive gradient method"
by S. Sra and D. Kim
Working draft of Jun. 2010;
Preprint [.pdf]
- "Tackling box-constrained convex optimization via a new projected quasi-Newton approach"
by D. Kim, S. Sra, I. S. Dhillon
in SIAM Journal on Scientific Computing (SISC) Oct. 2010;
Paper: [.pdf]
- "A scalable trust-region algorithm with application to mixed-norm regression"
by D. Kim, S. Sra, I. S. Dhillon
Interational Conference on Machine Learning (ICML) 2010;
Paper: [.pdf]
- "Multiframe Blind Deconvolution, Super-Resolution, and Saturation Correction via Incremental EM"
by S. Harmeling, S. Sra, M. Hirsch, B. Schölkopf
in IEEE International Conference on Image Processing (ICIP). 2010;
Paper: [.pdf]
- "Efficient Filter Flow for Space-Variant Multiframe Blind Deconvolution"
by M. Hirsch, S. Sra, B. Schölkopf, S. Harmeling
in IEEE Conf. Computer Vision & Pattern Recognition (CVPR 2010);
Paper: [.pdf]
- "Sparse nonnegative matrix approximation: new formulations and algorithms."
by R. Tandon and S. Sra.
MPI Technical Report #193. Sep. 2010;
Paper: [.pdf]
- "Fast algorithms for total-variation based optimization"
by A. J. Barbero and S. Sra.
MPI Technical Report #194. Aug. 2010;
Paper: [.pdf]
- "Generalized proximity and projection with norms and mixed-norms"
by S. Sra
MPI Technical Report #192. May 2010;
Paper: [.pdf]
2009
- "Efficient Filter Flow for Space-Variant Multiframe Blind Deconvolution"
by M. Hirsch, S. Sra, B. Schölkopf and S. Harmeling
MPI Technical Report #188, Nov. 2009;
Paper: [.pdf]
- "Text Clustering with Mixture of von Mises-Fisher Distributions"
by A. Banerjee, I. S. Dhillon, J. Ghosh, and S. Sra
in Text Mining: Theory, Applications, and Visualization
Book edited by: A. N. Srivastava and M. Sahami. CRC Press (2009);
(Invited chapter)
Preprint: [.pdf]
- "Approximation Algorithms for Tensor clustering"
by S. Jegelka, S. Sra, A. Banerjee
in Algorithmic Learning Theory (ALT) 2009.;
Extended version [cs.DS/0812.0389]
- "Online Blind Deconvolution for Astronomy"
by S. Harmeling, M. Hirsch, S. Sra, B. Schölkopf
in IEEE Interational Conferemce on Computational Photography (ICCP). 2009;
Paper: [.pdf]
- "A new non-monotonic algorithm for PET image reconstruction"
by S. Sra, D. Kim, I. S. Dhillon, B. Schölkopf
in IEEE Nuclear Science Symposium / Medical Imaging Conf. (NSS/MIC). 2009;
Conference Record M03-2: [.pdf]
- "Scalable Semidefinite Programming using Convex Perturbations"
by B. Kulis, S. Sra, I. S. Dhillon
in Artificial Intelligence and Statistics (AISTATS) 2009;
Paper: [.pdf]
2008
- "Block-Iterative Algorithms for Non-negative Matrix Approximation"
by S. Sra
in IEEE International Conference on Data Mining (ICDM) 2008;
Paper: [.pdf]
- "The Metric Nearness Problem"
by J. Brickell, I. S. Dhillon, S. Sra, J. A. Tropp
in SIAM J. Matrix Analysis and Applications vol. 30 no. 1 pp. 375-396 (2008)
SIAM Outstanding Paper Prize 2011 See: SIAM Website
(one of three papers selected out of all papers published in SIAM Journals
in the three years 2008--2010);
Paper: [.pdf]; Preprint: [.pdf]
- "Approximation Algorithms for Bregman Clustering Co-clustering and Tensor Clustering"
by S. Sra, S. Jegelka, and A. Banerjee
MPI Technical Report #177 2008;
Paper: [.pdf]
- "Block iterative algorithms for non-negative matrix approximation"
by S. Sra
MPI Technical Report #176 2008;
Paper: [.pdf]
- "Non-monotonic Poisson Likelihood Maximization"
by S. Sra, D. Kim, and B. Schölkopf;
MPI Technical Report #170, Jun. 2008
Paper: [.pdf]
- "A New Non-monotonic Gradient Projection Method for the Non-negative Least Squares Problem"
by D. Kim, S. Sra, and I. S. Dhillon
Computer Sciences, University of Texas at Austin, TR-08-28. Jun. 2008;
Paper: [-NA-]
2007
- "Matrix Nearness Problems in Data Mining"
by S. Sra
Ph.D. Thesis. University of Texas at Austin. Aug. 2007;
Thesis: [.pdf]
- "Information-theoretic Metric Learning"
by J. V. Davis, B. Kulis, P. Jain, S. Sra, I. S. Dhillon
in International Conference on Machine Learning (ICML) 2007.;
(Best Student Paper)
Paper: [.pdf]
- "Fast Newton-type Methods for the Least Squares Nonnegative Matrix Approximation Problem"
by D. Kim, S. Sra, I. S. Dhillon
SIAM International Conference on Data Mining (SDM) 2007;
Recognized within best of SDM 2007 papers
Paper: [.pdf]
- "Fast Projection-Based Methods for the Least Squares Nonnegative Matrix Approximation Problem"
by D. Kim, S. Sra, I. S. Dhillon
in Statistical Analysis and Data Mining vol. 1 pp. 38-51 (2007);
(Invited Paper)
Paper: [.pdf]
- "Scalable Semidefinite Programming using Convex Perturbations"
by B. Kulis, S. Sra, S. Jegelka, and I. S. Dhillon
Comp. Sci., Univ. of Texas at Austin, TR-07-47, Sep. 2007;
Paper: [.pdf]
- "A New Projected Quasi-Newton Approach for solving the Nonnegative Least-Squares Problem"
D. Kim, S. Sra, and I. S. Dhillon
Comp. Sci., Univ. of Texas at Austin, TR-06-54, May 2007;
Paper: [.pdf]
- "Modeling data using directional distributions: Part II"
S. Sra, P. Jain, and I. S. Dhillon
Comp. Sci., Univ. of Texas at Austin, TR-07-05, Feb. 2007;
Paper: [.pdf]
2006--2003
- "Information-theoretic Metric Learning"
by J. V. Davis, B. Kulis, S. Sra, and I. S. Dhillon
NIPS 2006 Workshop on learning to compare examples, Dec. 2006;
Paper: [.pdf]
- "Incremental Aspect Models for Mining Document Streams"
by A. Surendran, S. Sra
in Principles and Practice of Knowledge Discovery in Databases (PKDD) 2006;
Paper: [.pdf]
- "Efficient Large Scale Linear Programming Support Vector Machines"
by S. Sra
in European Conference on Machine Learning (ECML) 2006.;
Paper: [.pdf]
- "Row-action Methods for Compressed Sensing"
by S. Sra, J. A. Tropp
in International Conference on Acoustics, Speech, and Signal Processing (ICASSP) 2006.;
Paper: [.pdf]
- "Clustering on the Unit Hypersphere using von Mises-Fisher Distributions"
by A. Banerjee, I. S. Dhillon, J. Ghosh, S. Sra
in Journal of Machine Learning Research (JMLR) vol. 6 pp. 1345-1382 (2005);
Paper: [.pdf]
- "Generalized Nonnegative Matrix Approximations with Bregman Divergences"
by I. S. Dhillon, S. Sra
in Advances Neural Information Processing Systems (NIPS) 2005.;
Paper: [.pdf]
- "Minimum Sum Squared Residue based Co-clustering of Gene Expression data"
by H. Cho, I. S. Dhillon, Y. Guan, S. Sra
in SIAM International Conference on Data Mining (SDM) 2004;
Paper: [.pdf]
- "Triangle Fixing Algorithms for the Metric Nearness Problem"
by I. S. Dhillon, S. Sra, J. A. Tropp
in Advances in Neural Information Processing Systems (NIPS) 2004;
Paper: [.pdf]
- "Generative Model-Based Clustering of Directional Data"
by A. Banerjee, I. S. Dhillon, J. Ghosh, S. Sra
in International Conference on Knowledge Discovery and Data Mining (KDD) 2003.;
Paper: [.pdf]
- "Nonnegative Matrix Approximation: Algorithms and Applications"
by S. Sra and I. S. Dhillon
Comp. Sci., Univ. of Texas at Austin TR-06-27, Jun 2006;
Paper: [.pdf]
- "Generalized Nonnegative Matrix Approximations using Bregman Divergences"
by I. S. Dhillon and S. Sra
Comp. Sci., Univ. of Texas at Austin TR-05-31, Jun 2005;
Paper: [.pdf]
- "Triangle Fixing Algorithms for the Metric Nearness Problem"
by I. S. Dhillon, S. Sra, and J. A. Tropp
Comp. Sci., Univ. of Texas at Austin TR-04-22, Jun 2004;
Paper: [.pdf]
- "The Metric Nearness Problem with Applications"
by I. S. Dhillon, S. Sra, and J. A. Tropp
Comp. Sci., Univ. of Texas at Austin TR-03-23, July 2003;
Paper: [.ps.gz]
- "Expectation Maximization for Clustering on Hyperspheres"
by A. Banerjee, I. S. Dhillon, J. Ghosh, and S. Sra
Comp. Sci., Univ. of Texas at Austin TR-03-07, Feb. 2003;
Paper: [.ps.gz]
- "Modeling Data using Directional Distributions"
by I. S. Dhillon and S. Sra
Comp. Sci., Univ. of Texas at Austin TR-03-06, Jan. 2003;
Paper: [.ps.gz]