|
Research Area: Machine Learning |
Profile
Dr. Keerthi joined Yahoo! Research in December 2003. Over the last fifteen years his research has focused on the development of practical algorithms for a variety of areas, such as machine learning, robotics, computer graphics and optimal control.
His works on support vector machines (improved SMO algorithm), polytope distance computation (GJK algorithm) and model predictive control (stability theory) are highly cited.
He has published more than 100 papers in leading journals and conferences. His current research focuses on kernel methods, in particular on the development of fast kernel methods for large scale datamining problems.
Prior to joining Yahoo!, he worked for 10 years at the Indian Institute of Science, Bangalore, and for 5 years at the National University of Singapore.
Dr. Keerthi is an Associate Editor for the IEEE Transactions on Automation Science and Engineering.
Recent Publications, Projects and News
- Semi-Supervised Gaussian Process Classifiers. Sindhwani, Vikas ; Chu, Wei ; Keerthi, S. Sathiya, IJCAI, 2007
- Newton Methods for Fast Semisupervised Linear SVMs. Sindhwani, Vikas ; Keerthi, S. Sathiya, Large Scale Kernel Machines, MIT Press, 2007
- A Fast Tracking Algorithm for Generalized LARS/LASSO. Keerthi, S. Sathiya ; Shevade, Shirish K., IEEE Transactions on Neural Networks, 2007
- Fast Generalized Cross-Validation Algorithm for Sparse Model Learning. Sundararajan, S. ; Shevade, Shirish Krishnaj ; Keerthi, S. Sathiya, Neural Computation, 2007
- Support Vector Ordinal Regression. Chu, Wei ; Keerthi, S. Sathiya, Neural Computation, 2007
- An Efficient Method for Gradient-Based Adaptation of Hyperparameters in SVM Models Keerthi, S. Sathiya ; Sindhwani, Vikas ; Chapelle, Olivier, Advances in Neural Information Processing Systems 19, MIT Press, 2007
- Relational Learning with Gaussian Processes Chu, Wei ; Sindhwani, Vikas ; Ghahramani, Zoubin ; Keerthi, S. Sathiya, Advances in Neural Information Processing Systems 19, MIT Press, 2007
- Branch and Bound for Semi-Supervised Support Vector Machines Chapelle, Olivier ; Sindhwani, Vikas ; Keerthi, S. Sathiya, Advances in Neural Information Processing Systems 19, MIT Press, 2007
- Deterministic annealing for semi-supervised kernel machines. Sindhwani, Vikas ; Keerthi, S. Sathiya ; Chapelle, Olivier, ICML, 2006
- Large scale semi-supervised linear SVMs. Sindhwani, Vikas ; Keerthi, S. Sathiya, SIGIR, 2006
- Building Support Vector Machines with Reduced Classifier Complexity. Keerthi, S. Sathiya ; Chapelle, Olivier ; DeCoste, Dennis, Journal of Machine Learning Research, 2006
- Developing parallel sequential minimal optimization for fast training support vector machine. Cao, L. J. ; Keerthi, S. Sathiya ; Ong, Chong Jin ; Uvaraj, P. ; Fu, X. J. ; Lee, H. P., Neurocomputing, 2006
- New approaches to support vector ordinal regression. Chu, Wei ; Keerthi, S. Sathiya, ICML, 2005
- An improved conjugate gradient scheme to the solution of least squares SVM. Chu, Wei ; Keerthi, S. Sathiya ; Ong, Chong-Jin, IEEE Transactions on Neural Networks, 2005
- Generalized LARS as an effective feature selection tool for text classification with SVMs. Keerthi, S. Sathiya, ICML, 2005
- Which Is the Best Multiclass SVM Method? An Empirical Study. Duan, Kaibo ; Keerthi, S. Sathiya, Multiple Classifier Systems, 2005
- A matching pursuit approach to sparse Gaussian process regression. Keerthi, S. Sathiya ; Chu, Wei, NIPS, 2005
- A Modified Finite Newton Method for Fast Solution of Large Scale Linear SVMs. Keerthi, S. Sathiya ; DeCoste, Dennis, Journal of Machine Learning Research, 2005
- A Fast Dual Algorithm for Kernel Logistic Regression. Keerthi, S. Sathiya ; Duan, Kaibo ; Shevade, S. ; Poo, A., Machine Learning, 2005
- Predictive Approaches for Sparse Model Learning. Shevade, Shirish Krishnaj ; Sundararajan, S. ; Keerthi, S. Sathiya, ICONIP, 2004