Latest News and Events

The SAMSI-FODAVA Workshop on Interactive Visualization and Analysis of Massive Data will be held on December 10-12, 2012.
Posted: October 02, 2012
The FODAVA Annual Meeting will immediately follow (Dec 12-13) the SAMSI/FODAVA joint workshop at the same location.
Posted: September 05, 2012
Many of the modern data sets such as text and image data can be represented in high-dimensional vector spaces and have benefited from computational methods that utilize advanced techniques from num
Posted: June 30, 2012

Forum on Geometric Aspects of Machine Learning and Visual Analytics: Recent Developments and Future Challenges

The primary aim of the Forum is to bring together researchers in Computer Science, Mathematics, Statistics and related areas working on geometric problems in Machine Learning with a potential impact in Data and Visual Analytics.
In the recent years, there has been significant progress in Machine and Statistical Learning in general, the design of algorithms that extract and process information from data sets, and the mathematical understanding of the limits and capabilities of such algorithms.

In this forum we will focus on recent trends in Machine Learning that aim at understanding the geometric nature of Machine Learning problems. It has been understood that there are rather subtle geometric structures involved in complex high dimensional data sets that have to be revealed in the process of their analysis and visualization. These structures are often hidden even in the data sets that seemingly have nothing to do with geometry (such data sets are common in many Visual Analytics applications). Novel techniques, theoretical insights, algorithms and computational techniques have been developed along this lines and will be discussed in the forum.


V. Koltchinskii, School of Mathematics, Georgia Institute of Technology
M. Maggioni, Department of Mathematics, Duke University
H. Park, Division of Computational Science and Engineering, Georgia Institute of Technology
A. Varshney, Department of Computer Science, University of Maryland

Dates: October 11-12, 2009
Location: IEEE VisWeek, Atlantic City, New Jersey
Online Registration
Hotel Registration


October 11, Dennis A&B, Ballys Hotel: Click on the titles to see each speaker's abstract.


Session 1: Chair: Haesun Park

0830 - 0900 Thomas, Jim (Pacific Northwest National Laboratory) - The Opportunities and Challenges within Visual Analytics?

0900 - 0940 Carlson, Gunnar (Stanford University) - Partial clustering and visualization

0940 - 1010  Lerman, Gilad (University of Minnesota, Twin Cities) - Foundations of Multi-Manifold Modeling Algorithms

Session 2: Chair: Torsten Möller

1030 - 1100 Vidal, Rene (The Johns Hopkins University) - Sparse Subspace Clustering

1100 - 1130  Rigollet, Philippe (Princeton University) - Nonparametric Bandits with Covariates

1130 - 1200  Scott, Clayton (Michigan University) - Algorithms and a Visual Interface to Support Query Learning Under Uncertainty

1200 - 0130  Lunch break

Session 3: Chair: Vladimir Koltchinskii

0130 - 0200 Mukherjee, Sayan (Duke University) - Visualizing Conditional Dependencies in Hyper-Graph Models

0200 - 0230  Belkin, Misha (Ohio State University) - Towards Understanding Mixtures of Gaussians: Spectral Methods and Polynomial Time Learning with No Separation

0230 - 0300 Jebara, Tony (Columbia University) - Learning from Data using Matchings and Graphs

Session 4: Chair: Amitabh Varshney

0320 - 0400 Vempala, Santosh (Georgia Institute of Technology) - Affine-invariant Principal Components

0400 - 0430  Whitaker, Ross (University of Utah) - Parameterizing high-dimensional data sets with kernel map manifolds

0430 - 0500  Keim, Daniel (University of Konstanz, Germany) - Visual Analytics Techniques for Clustering of High-dimensional Data

0500 - 0530  Ma, Kwan-Liu (University of California, Davis) - Uncertainty-Aware Visualization of Networks


October 12, Traymore A, Ballys Hotel: 

0900 - 1200  Report writing session: Recent Developments and Future Challenges in Visual Analytics in relation to Geometric Aspects of Machine Learning