AML08: Algebraic Methods in Machine Learning
Symposium and Workshop at NIPS'08 December 11-12, Vancouver/Whistler


Organizers:


 Risi Kondor, Gatsby Unit, UCL
Guy Lebanon, CSE, Georgia Tech
Jason Morton, Math, Stanford
Contact: amlworkshop08 at gmail dot com
Bulletin board: [here] Please write to the above address if you wish to post something on this board.
Summary: There has recently been a surge of interest in algebraic methods in machine learning. In no particular order, this includes: new approaches to ranking problems e.g., [1][2][7][7b][12][13][14][17]; the budding field of algebraic statistics e.g., [3][4][15][16]; and various applications of non-commutative Fourier transforms [6][8][9][10]. The aim of the workshop is to bring together these distinct communities, explore connections, and showcase algebraic methods to the machine learning community at large.
AML'08 is intended to be accessible to researchers with no prior exposure to abstract algebra. The program includes three short tutorials that will cover the basic concepts necessary for understanding cutting edge research in the field.
Registration: AML'08 is part of the NIPS'08 conference and workshops. To attend the symposium you must register either for the NIPS conference or the NIPS workshops. To attend the workshop, you must register for the workshops. For more information on registration, please see the NIPS website.
Preliminary program:
Symposium session
Thursday, Dec. 11, Regency Ballroom A/B at the Vancouver Hyatt [abstracts]
13.30 - 14.00 Risi Kondor: Non-commutative harmonic analysis [slides]
14.05 - 14.35 Guy Lebanon: Modeling distributions on permutations and partial ranking [slides]
14.40 - 15.10 Jason Morton: Algebraic models for multilinear dependence [slides]
15.15 - 15.25 coffee break
15.25 - 15.55 Yanxi Liu: Symmetry Group-based Learning for Regularity Discovery from Real World Patterns
16.00 - 16.30 Marina Meila: Estimation and model selection in stagewise ranking: a representation story [slides]

Workshop session 1
Friday, Dec. 12, Callaghan Room at the Whistler Westin [abstracts]
7.30 - 8.00 Stephen E. Fienberg: Algebraic statistics for random graph models: Markov bases and their uses [slides] [video]
8.05 - 8.35 Adrian Dobra: Algebraic statistics and contingency tables [slides] [video]
8.40 - 9.10 Keisuke Yamazaki: Toric Modification on Mixture Models [slides] [video]
9.15 - 9.25 coffee break
9.25 - 9.55 Vincent Auvray: Learning Parameters in Discrete Naive Bayes Models by Computing Fibers of the Parametrization map [slides] [video]
10.00 - 10.30 Paul von Bunau: Stationary Subspace Analysis [slides] [video]

Workshop session 2
Friday, Dec. 12, Callaghan Room at the Whistler Westin [abstracts]
15.30 - 16.00 Doru Balcan: Alternatives to the Discrete Fourier Transform [slides] [video]
16.05 - 16.35 Lek-Heng Lim: Graph Helmholtzian and rank learning [slides] [video]
16.40 - 17.10 Xiaoye Jiang: Identity Management On Homogeneous spaces [slides] [video]
17.15-17.25 coffee break
17.25 - 17.55 Jonathan Huang: Exploiting Probabilistic Independence for Permutations [slides] [video]
17.55 - 18.30 Tiberio Caetano: Consistent structured estimation for weighted bipartite matching [slides] [video]

References:
[1]N. Ailon: Aggregation of partial rankings, p-rakings and top-m lists (SODA 2007) [pdf]
[2]N. Ailon and M. Mohri: An efficient reduction of ranking to classification (COLT 2008) [pdf]
[3]P. Diaconis and B. Sturmfels: Algebraic algorithms for sampling from conditional distributions (Annals of Statistics, 26 1:363-97) [pdf]
[4]M. Drton, T. Richardson: Graphical methods for efficient likelihood inference in Gaussian covariance models (JMLR) [pdf]
[5]D. Geiger, C. Meek and B. Sturmfels: On the toric algebra of graphical models (Annals of Statistics 34 (2006) 1463-1492) [pdf]
[6]J. Huang, C. Guestrin and L. Guibas: Efficient inference for distributions on permutations (NIPS 2007) [pdf]
[7]S. Jagabathula, D. Shah: Inferring rankings under constrained sensing (NIPS 2008)
[7b]X. Jiang, L-H Lim, Y. Yao, Y. Ye: Learning to rank with combinatorial Hodge theory (arXiv) [pdf]
[8]R. Kondor: A novel set of translation and rotation invariant features for images based on the non-commutative bispectrum (arXiv) [pdf]
[9]R. Kondor, A. Howard and T. Jebara: Multi-object tracking with representations of the symmetric group (AISTATS 2007) [pdf]
[10]R. Kondor and K. M. Borgwardt: The skew spectrum of graphs (ICML 2008) [pdf]
[11]J. M. Landsberg, J. Morton: The Geometry of Tensors: Applications to complexity, statistics and engineering (in preparation)
[12]G. Lebanon and J. Lafferty: Conditional models on the ranking poset (NIPS 2002) [pdf]
[13]G. Lebanon and Y. Mao: Non-parametric modeling of partially ranked data (NIPS 2007) [pdf]
[14]M. Meila, K. Phadnis, A. Patterson and J. Blimes: Consensus ranking under the exponential model (tech report, 2007) [pdf]
[15]J. Morton: Relations among conditional probabilities (arXiv) [pdf]
[16]S. Sullivant: Algebraic geometry of Gaussian Bayesian networks (to appear in Advances in Applied Mathematics, arXiv) [pdf]
[17]M. Warmuth: Learning permutations with exponential weights (COLT 2007) [pdf]