Thursday, July 01, 2010

CS: Presentations made at Sparsity and Computation workshop and at the 2010 Modern Massive Data Sets

You thought you'd go home for the week-end without much to read on your new shiny iPad, well forget that. Today two workshops have released some of the slides presentation made there. To get full publicity for the event, for their sponsors or for the presenters, the process of getting slides from the presenters and making them available should not be an after thought. Enjoy!

The workshop on Sparsity and Computation at the Hausdorff-center for Mathematics took place three weeks ago and it looks like we now have access to some of the presentations made there. Here there are:

Programme and Slides

Please click on the title to obtain the pdf-file of the talk (if available).

Monday, June 7, 2010

8:30-9:00 Registration

9:00-9:10 Opening

9:10-10:00 Ingrid Daubechies, Princeton University
Animation, Teeth and Skeletons ...
10:10-11:00 Erich Novak, University of Jena
How can we obtain tractability of multivariate problems?
11:00-11:30 Coffee Break

11:30-12:00 Gitta Kutyniok, University of Osnabrück
Geometric Separation by Single-Pass Alternating Thresholding
12:00-12:30 Lawrence Carin, Duke University
Statistical nonlinear matrix completion
12:30-15:00 Lunch Break

15:00-15:30 Wolfgang Hackbusch, MPI Leipzig
1D data compression by tensor based methods
15:30-16:00 Rob Stevenson, KdVI Amsterdam
Space-Time Adaptive Wavelet Methods for Parabolic Evolution Problems
16:00-16:30 Angela Kunoth, University of Paderborn
Space-Time Adaptive Wavelet Methods for Control Problems Constrained by Parabolic PDEs
16:30-17:00 Coffee Break

17:00-17:30 Stephan Dahlke, University of Marburg
Multilevel Preconditioning and Adaptive Sparse Solution of Inverse Problems
17:30-18:00 Gerd Teschke, Hochschule Neubrandenburg
Sparse recovery and compressed sensing in inverse problems
18:15 Reception

Tuesday, June 8, 2010

10:00-10:50 Emmanuel Candès, Stanford University
Robust Principal Component Analysis?
10:50-11:30 Coffee Break

11:30-12:00 Benjamin Recht, University of Wisconsin
The Convex Geometry of Inverse Problems
12:00-12:30 Rachel Ward, New York University
Sparse Legendre expansions via l1 minimization
12:30-15:00 Lunch Break

15:00-15:30 Stefan Kunis, Technical University of Chemnitz & Helmholtz Center Munich
Sparse and Fast Fourier Transforms in Biomedical Imaging
15:30-16:00 Yonina Eldar, Technion Haifa
Xampling: Analog-to-digital at Sub-Nyquist rates
16:00-16:30 Mauro Maggioni, Duke University
Multiscale Geometric Analysis of Data Sets
16:30-17:00 Coffee Break

17:00-17:30 Przemyslaw Wojtaszczyk, University of Warsaw
Approximation of functions of few variables in high dimensions
17:30-18:00 Volodya Temlyakov, University of South Carolina
Orthogonal Super Greedy Algorithm and Applications in Compressed Sensing

Wednesday, June 9, 2010

9:00-9:50 Jean-Luc Guermond, Texas A&M University
Entropy viscosity for nonlinear conservation laws
10:00-10:50 Alain Pajor, Université Paris Est - Marne-la-Vallée
Random polytopes and neighborliness
10:50-11:30 Coffee Break

11:30-12:00 Nicole Tomczak-Jaegermann, University of Alberta
On random matrices with independent log-concave columns
12:00-12:30 Bojan Popov, Texas A&M University
Nonlinear Approximation Techniques Using L1
12:30-14:00 Lunch Break

14:00-14:50 Joel Tropp, California Institute of Technology
User-Friendly Tail Bounds for Sums of Random Matrices
15:15 Excursion

Thursday, June 10, 2010

9:00-9:50 Wolfgang Dahmen, RWTH Aachen
Convergence Rates for Greedy Algorithms in Reduced Basis Methods
10:00-10:50 Christoph Schwab, ETH Zurich
Convergence Rates for Sparse Adaptive Tensor Approximations of parametric and stochastic PDEs
10:50-11:30 Coffee Break

11:30-12:20 Albert Cohen, Université Pierre et Marie Curis Paris
Analysis of the reduced basis method for parametric elliptic PDEs
12:20-15:00 Lunch Break

15:00-15:30 Mark Iwen, University of Minnesota
Sparse Fourier Approximation in High Dimensions
15:30-16:00 Karin Schnass, RICAM Linz
Dictionary Identification - Sparse Matrix-Factorisation via l1-Minimisation
16:00-16:30 Maryam Fazel, University of Washington
A nullspace approach to low-rank matrix recovery
16:30-17:00 Coffee Break

17:00-17:30 Andrea Montanari, Stanford University
Message passing algorithms, random convex problems, and the risk of the LASSO
17:30-18:00 Michael Elad, Technion Haifa
Topics in Minimum-Mean-Squared-Error (MMSE) Estimation in Sparse Approximation
20:30 Dinner at restaurant Im Stiefel (Bonngasse 30)

Friday, June 11, 2010

9:00-9:50 Piotr Indyk, MIT
Sparse Recovery for Earth Mover Distance
10:00-10:30 Thomas Blumensath, University of Southhampton
Three Generalisations of Compressed Sensing
10:30-11:00 Özgür Yilmaz, University of British Columbia
Quantization of Compressed Sensing Measurements
11:00-11:30 Coffee Break

11:30-12:00 Justin Romberg, Georgia Institute of Technology
Random coding for forward modeling
12:00-12:30 Gabriele Steidl, University of Mannheim
Dithering by Differences of Convex Functions
12:30-14:40 Lunch Break

14:40-15:30 Michael Griebel, University of Bonn
Dimension-wise integration of high-dimensional functions with applications to finance
15:30-16:15 Coffee Break

16:15-17:15 Roman Vershynin, University of Michigan
Jointly with the Bonn Mathematical Colloquium
Non-asymptotic theory of random matrices and sparsity



The workshop on the 2010 Modern Massive Data Sets took place two weeks ago at Stanford. The site has a list of the talks but all the links link to a placeholder. Some talks are available however and so I am featuring only those talks that have an actual presentation attached to them. They can be found here as well.

Tuesday, June 15, 2010. Theme: Large-scale Data and Large-scale Computation

Time Talk
8:00 - 10:00 Breakfast and Registration -- outside Cubberley Auditorium (at the Stanford School of Education, just off the Main Quad)
9:45 - 10:00 Welcome and Opening Remarks -- in Cubberley Auditorium
10:00 - 11:00 Tutorial: Peter Norvig
Internet-Scale Data Analysis
11:00 - 11:30 Ashok Srivastava
Virtual Sensors and Large-Scale Gaussian Processes
11:30 - 12:00 John Langford
A method for Parallel Online Learning
2:00 - 3:00 Tutorial: John Gilbert
Combinatorial Scientific Computing: Experience and Challenges
3:00 - 3:30 Deepak Agarwal
Estimating Rates of Rare Events through Multiple Hierarchies
3:30 - 4:00 James Demmel
Minimizing Communication in Linear Algebra
4:30 - 5:00 Dmitri Krioukov
Hyperbolic mapping of complex networks
5:00 - 5:30 Mehryar Mohri
Matrix approximation for large-scale learning
5:30 - 6:00 David Bader
Massive Scale Analytics of Streaming Social Networks
6:00 - 6:30 Ely Porat
Fast Pseudo-Random Fingerprints

Wednesday, June 16, 2010. Theme: Networked Data and Algorithmic Tools

Time Talk
9:00 - 10:00 Tutorial: Peter Bickel
Statistical Inference for Networks
10:00 - 10:30 Jure Leskovec
Inferring Networks of Diffusion and Influence
11:00 - 11:30 Michael W. Mahoney
Geometric Network Analysis Tools
11:30 - 12:00 Edward Chang
AdHEat - A New Influence-based Social Ads Model and its Tera-Scale Algorithms
12:00 - 12:30 Mauro Maggioni
Intrinsic dimensionality estimation and multiscale geometry of data sets
2:30 - 3:00 Guillermo Sapiro
Structured Sparse Models
3:00 - 3:30 Alekh Agarwal and Peter Bartlett
Information-theoretic lower bounds on the oracle complexity of convex optimization
3:30 - 4:00 John Duchi and Yoram Singer
Composite Objective Optimization and Learning for Massive Datasets
4:30 - 5:00 Steven Hillion
MAD Analytics in Practice
5:00 - 5:30 Matthew Harding
Outlier detection in financial trading networks
5:30 - 6:00 Neel Sundrahan
Large Dataset Problems at the Long Tail

Thursday, June 17, 2010. Theme: Spectral Methods and Sparse Matrix Methods

Time Talk
9:00 - 10:00 Tutorial: Sebastiano Vigna
Spectral Ranking
10:00 - 10:30 Robert Stine
Streaming Feature Selection
11:00 - 11:30 Konstantin Mischaikow
A combinatorial framework for nonlinear dynamics
11:30 - 12:00 Alfred Hero
Sparse correlation screening in high dimension Sparse correlation screening in high dimension
12:00 - 12:30 Susan Holmes
Challenges in Statistical Analyses: Heterogeneous Data
2:30 - 3:30 Tutorial: Piotr Indyk
Sparse Recovery Using Sparse Matrices
3:30 - 4:00 Sayan Mukherjee
Efficient dimension reduction on massive data
4:30 - 5:00 Padhraic Smyth
Statistical Modeling of Large-Scale Sensor Count Data
5:00 - 5:30 Ping Li
Compressed Counting for Data Stream Computation and Entropy Estimation
5:30 - 6:00 Edo Liberty
Scaleable Correlation Clustering Algorithms

Friday, June 18, 2010. Theme: Randomized Algorithms for Data

Time Talk
9:00 - 10:00 Tutorial: Petros Drineas
Randomized Algorithms in Linear Algebra and Large Data Applications
10:00 - 10:30 Gunnar Martinsson
Randomized methods for computing the SVD/PCA of very large matrices
11:00 - 11:30 Ilse Ipsen
Numerical reliability of randomized algorithms
11:30 - 12:00 Patrick Wolfe
Randomized Algorithms and Sampling Schemes for Large Matrices
12:00 - 12:30 Alexandre d'Aspremont
Subsampling, Spectral Methods & Semidefinite Programming
2:30 - 3:00 Gary Miller
Specialized System Solvers for very large Systems: Theory and Practice
3:00 - 3:30 John Wright and Emmanuel Candes
Robust Principal Component Analysis?
3:30 - 4:00 Alon Orlitsky
Estimation, Prediction, and Classification over Large Alphabets
4:30 - 5:00 Ken Clarkson
Numerical Linear Algebra in the Streaming Model
5:00 - 5:30 David Woodruff
Fast Lp Regression in Data Streams
Credit: Presentation of Justin Romberg.

No comments:

Printfriendly