Recently, linear algebra techniques have given a fundamentally different perspective for learning and inference in latent variable models. Exploiting the underlying spectral properties of the model parameters has led to fast, provably consistent methods for structure and parameter learning that stand in contrast to previous approaches, such as Expectation Maximization, which suffer from local optima and
slow convergence. Furthermore, these techniques have given insight into the nature of latent variable models.

In this workshop, via a mix of contributed/invited talks, posters, and discussion, we seek to explore the theoretical and applied aspects of spectral methods including the following major themes:

(1) How can spectral techniques help us develop fast and local minima free solutions to real world problems involving latent variables in natural language processing, dynamical systems, computer vision etc. where existing methods such as Expectation Maximization are unsatisfactory?

(2) How can these approaches lead to a deeper understanding and interpretation of the complexity of latent variable models?

We welcome the submission of papers related to the intersection of spectral/linear algebra methods and probabilistic modeling. Possible topics could include (but are not limited to) spectral algorithms for learning and inference in probabilistic models, tensor decomposition methods, theoretical results, and applications of these approaches.
The goal of the workshop is to bring together a wide variety of researchers from both theoretical and applied areas, and thus is open to a wide range of papers.

-----Submission------
Submitted papers should be in NIPS 2012 format with a maximum of 4 pages (not including references). Please email your submission to nips2012spectrallearning@gmail.com.

Sept 23, 2012 - Deadline for submission
Oct 7, 2012 - Notification of acceptance

If you have any questions, feel free to contact us at
nips2012spectrallearning@gmail.com.

--------Organizers------
Ankur Parikh (Carnegie Mellon)
Le Song (Georgia Tech)
Eric Xing (Carnegie Mellon)