|Kristin P. Bennett
Professor, Department of Mathematical Sciences
Rensselaer Polytechnic Institute
Ph.D., University of Wisconsin, Madison, 1993
- Combining operations research and artificial intelligence problem solving methods.
- Mathematical programming approaches to problems in artificial intelligence such as machine learning, neural networks, pattern recognition, and planning.
- Application of these techniques to medical, financial and scientific problems.
- Adaptation of these algorithms for parallel machines.
- Mathematical programming approaches to other areas in computer sciences such as genetic algorithms and database query optimization.
M. Momma and K. Bennett, “Constructing Orthogonal Latent Features for Arbitrary Loss”, Feature Extraction, Foundations and Applications, Isabelle Guyon, Steve Gunn, Masoud Nikravesh, and Lofti Zadeh, editors, Springer, to appear, accepted 2004.
“Regression Error Characteristic Curves ”, Jinbo Bi and K. P. Bennett, Proceedings of the 20th International Conference on Machine Learning, 2003.
“A Geometric Approach to Support Vector Regression”, Jinbo Bi and K. P. Bennett, Neurocomputing, 55, 2003, pp. 79-108.
Sparse Kernel Partial Least Squares Regression. M. Momma and K.P. Bennett, Proceedings of Conference on Learning Theory, 2003.
K. P. Bennett and M. J. Embrechts, “An Optimization Perspective on Partial Least Squares”, in J.A.K. Suykens, G. Horvath, S. Basu, C. Micchelli, J. Vandewalle (Eds.) Advances in Learning Theory: Methods, Models and Applications, NATO Science Series III: Computer \& Systems Sciences, Volume 190, IOS Press Amsterdam, 2003,p. 227-250.
J. Bi, K. Bennett, M. Embrechts, C. Breneman, and M. Song, “Dimensionality Reduction via Sparse Support Vector Machines”, Journal of Machine Learning Research, 3, 2003, 1229-1243.
M. Song, C. Breneman, J. Bi, N. Sukumar, K. Bennett, S. Cramer, and N. Tugcu, “Prediction of Protein Retention Times in Anion-Exchange Chromatography Systems Using Support Vector Regression”, September 2002, Journal of Chemical Information and Computer Sciences.
K. Bennett, M. Momma, and J. Embrechts, MARK: A Boosting Algorithm for Heterogeneous Kernel Models, Proceedings of SIGKDD International Conference on Knowledge Discovery and Data Mining, 2002.
“Exploiting Unlabeled Data in Ensemble Methods”, with A. Demiriz, and R. Maclin, to appear in Proceedings of SIGKDD International Conference on Knowledge Discovery and Data Mining, 2002.
C. Breneman, K. Bennett, M. Embrechts, S. Cramer, M. Song, and J. Bi, “Descriptor Generation, Selection and Model Building in Quantitative Structure-Property Analysis”, in Chapter 11 of Experimental Design for Combinatorial and High Throughput Materials Development, J. Crawse editor, Wiley, 2002.
“A Pattern Search Method for Model Selection of Support Vector Regression with M. Momma, to appear in Proceedings of SIAM Conference on Data Mining, 2002.
”Duality, Geometry, and Support Vector Regression with Jinbo Bi, June 2001, Advances in Neural Information Processing Systems 14, T. Dietterich, S. Becker and Z. Ghahramani editors, MIT Press, Cambridge, pg 593-600. 2002.
Support Vector Machines: Hype or Hallelujah? with Colin Campbell, SIGKDD Explorations, 2,2, 2000, 1-13.
Support Vector Machine Regression in Chemometrics” with A. Demiriz, C. Breneman, M. Embrechts, “ Computing Science and Statistics, 2001
Sparse regression ensembles in infinite and finite hypothesis spaces, with Gunnar Raetsch and Ayhan Demiriz, Machine Learning, 48, 1-3, 2002, pp 193-221.
Linear Programming Boosting via Column Generation, with Ayhan Demiriz and John Shawe-Taylor, Machine Learning, 46:1, 2001, 225-254.
Constrained K-Means Clustering, with Paul Bradley and Ayhan Demiriz, Microsoft Research Technical Report 2000-65, May 2000.
A Column Generation Algorithm for Boosting , with Ayhan Demiriz and John Shawe-Taylor, Proceedings of the Seventeenth International Conference on Machine Learning, Pat Langley Editor, Morgan Kaufmann, San Francisco, 2000, pp. 65-72.
Duality and Geometry in SVM Classifiers, with Erin Bredensteiner, Proceedings of the Seventeenth International Conference on Machine Learning, Pat Langley Editor, Morgan Kaufmann, San Francisco, 2000, pp. 57-64.
Large Margin Trees for Induction and Transduction, with Donghui Wu, Nello Cristianini, and John Shawe-Taylor, ICML’99 Proceedings.
Density-Based Indexing for Approximate Nearest Neighbor Queries, with Usama Fayyad and Dan Geiger, Microsoft Research Technical Report 98-58, Microsoft Research, Redmond WA, 19998. Longer version of paper in KDD’99 Proceedings.
Enlarging Margins in Perceptron Decision Trees , with Donghui Wu, Nello Cristianini, and John Shawe-Taylor Machine Learning, 41:3, 295-313.
Semi-Supervised Clustering Using Genetic Algorithms, with Ayhan Demiriz and Mark Embrechts, ANNIE’99 (Artificial Neural Networks in Engineering), November 1999. Long version: Genetic Algorithm Approach for Semi-supervised Clustering is under review for journal publication.
Optimization Approaches to Semi-Supervised Learning , with Ayhan Demiriz, to appear in Applications and Algorithms of Complementarity, M. C. Ferris, O. L. Mangasarian, and J. S. Pang, editors. Kluwer Academic Publishers. Boston 2000.
Multicategory Classification by Support Vector Machines. with E. J. Bredensteiner, Computational Optimizations and Applications, 12, 1999, pp 53-79.
On Mathematical Programming Methods and Support Vector Machines, in Advances in Kernel Methods -- Support Vector Machines, A. Schoelkopf, C. Burges, and A. Smola, editors, MIT Press, Cambridge, MA, 1999, pp 307-326.
Semi-Supervised Support Vector Machines with A. Demiriz, Advances in Neural Information Processing Systems, 12, M. S. Kearns, S. A. Solla, D. A. Cohn, editors, MIT Press, Cambridge, MA, 1998, pp 368-374.
On Support Vector Decision Trees for Database Marketing with D. H. Wu and L. Auslender, R.P.I Math Report No. 98-100, Rensselaer Polytechnic Institute, Troy, NY, 1998. Long version of IJCNN’99 paper
A Support Vector Machine Approach to Decision Trees with J. Blue, R.P.I Math Report No. 97-100, Rensselaer Polytechnic Institute, Troy, NY, 1997.
Optimal Decision Trees with J. Blue R.P.I Math Report No. 214, 1996.
Geometry in Learning with E. Bredensteiner, Web Manuscript http:\\www.rpi.edu\~bennek\geometry2.ps September 1996. also in Geometry at Work, C. Gorini editors, Mathematical Association of America, Washington, D.C, 2000, 132-145.
Hybrid Extreme Point Tabu Search with J. Blue, R.P.I Math Report No. 240, 1996, appeared in European Journal of Operation Research 106, 1998, 676-688
An Extreme Point Tabu Search Method for Data Mining. with J. Blue, R.P.I Math Report No. 228, 1996.
Feature Minimization within Decision Trees. with E. Bredensteiner, Computational Optimizations and Applications, 10:2, 1998. 111-126
A Parametric Optimization Method for Machine Learning. with E. J. Bredensteiner, R.P.I Math Report No. 217, 1995. Final version appeared in INFORMS Journal on Computing, 9:3, pp. 311-318, 1997.
Global Tree Optimization: A Non-Greedy Decision Tree Algorithm. Computing Science and Statistics, 26, pp. 156-160, 1994
Serial and Parallel Multicategory Discrimination, SIAM Journal on Optimization, with O. L. Mangasarian, 4:4, pp. 722-734, 1994.
Multicategory Discrimination via Linear Programming, with O. L. Mangasarian, Optimization Methods and Software 3, pp. 27-39, 1993.
Bilinear Separation of Two Sets in n-Space, with O. L. Mangasarian, Computational Optimization and Applications, 2, pp. 207-227, 1993.
Decision Tree Construction Via Linear Programming, Proceedings of the 4th Midwest Artificial Intelligence and Cognitive Science Society Conference, Utica, Illinois, pp. 97-101, 1992.
Robust Linear Programming Discrimination of Two Linearly Inseparable Sets. with O. L. Mangasarian, Optimization Methods and Software, 1, pp. 23-34, 1992.
Neural Network Training Via Linear Programming, with O. L. Mangasarian, in P. M. Pardalos (ed.), Advances in Optimization and Parallel Computing, pp. 56-67, 1992.
Note that this research was based partially upon work supported by the National Science Foundation under Grant No. 970923 and No. 9979860.
Kristin P. Bennett