ABSTRACT
Aiming at the large memory footprint of traditional vector machine (Relevance Vector Machine, RVM) when processing big data in supervised learning, the idea of incremental learning is introduced into the traditional RVM and the incremental learning of RVM based on sparse model is studied. Method of an incremental learning algorithm of RVM based on sparse model is proposed. The algorithm considers the influence of the existing model and the new sample on the sparse RVM model, and transforms each incremental learning to the problem of solving the maximized edge likelihood function. The sparse RVM model is updated by solving the optimization problem continuously. Simulation results show that this method can effectively reduce the memory space requirements.
- Tipping M E. Tipping, M.E.: Sparse bayesian learning and the relevance vector machine. JMLR 1{J}. 2001, 1:211--244. Google ScholarDigital Library
- Boser B E, Guyon I M, Vapnik V N. A Training Algorithm for Optimal Margin Classifier{J}. Proceedings of Annual Acm Workshop on Computational Learning Theory, 2008, 5:144--152. Google ScholarDigital Library
- Bishop C M, Tipping M E. Variational relevance vector machines{C}// Sixteenth Conference on Uncertainty in Artificial Intelligence. Morgan Kaufmann Publishers Inc. 2000:46--53. Google ScholarDigital Library
- Chen H, Tino P, Yao X. Efficient probabilistic classification vector machine with incremental basis function selection{J}. IEEE Transactions on Neural Networks & Learning Systems, 2014, 25(2):356.Google ScholarCross Ref
- R&Xfc S, ping. Incremental Learning with Support Vector Machines{J}. Technical Reports, 2001, 228(4):641--642.Google Scholar
- Liu D, Zhou J, Pan D, et al. Lithium-ion battery remaining useful life estimation with an optimized Relevance Vector Machine algorithm with incremental learning{J}. Measurement, 2015, 63:143--151.Google ScholarCross Ref
- Burr T. Pattern Recognition and Machine Learning{M}// Pattern recognition and machine learning. Springer, 2006:461--462.Google Scholar
- Blekas K, Likas A. Sparse regression mixture modeling with the multi-kernel relevance vector machine{J}. Knowledge & Information Systems, 2014, 39(2):241--264. Google ScholarDigital Library
- Faul A, Avenuse J J T. Fast Marginal Likelihood Maximisation for Sparse Bayesian Models{J}. Proceedings of the Ninth International Workshop on Artificial Intelligence & Statistics, 2003:3--6.Google Scholar
- Tzikas D, Likas A, Galatsanos N. Incremental Relevance Vector Machine with Kernel Learning{C}// Artificial Intelligence: Theories, MODELS and Applications, Hellenic Conference on Ai, Setn 2008, Syros, Greece, October 2-4, 2008. Proceedings. DBLP, 2008:301--312. Google ScholarDigital Library
- http://archive.ics.uci.edu/ml/datasets.htmlGoogle Scholar
Index Terms
- Incremental Learning Algorithm Based on Relevance Vector Machine
Recommendations
A Fast Incremental Learning Algorithm Based on Twin Support Vector Machine
ISCID '14: Proceedings of the 2014 Seventh International Symposium on Computational Intelligence and Design - Volume 02Twin support vector machine is a novel classifier, it construct two nonparallel hyper planes instead of a single hyper plane to obtain four times faster than the usual SVM. With the result of traditional incremental learning method of SVM, we analyze ...
Incremental granular relevance vector machine
This paper focuses on extending the capabilities of relevance vector machine which is a probabilistic, sparse, and linearly parameterized classifier. It has been shown that both relevance vector machine and support vector machine have similar ...
A New Incremental Learning Support Vector Machine
AICI '09: Proceedings of the 2009 International Conference on Artificial Intelligence and Computational Intelligence - Volume 01A new incremental learning method for support vector machine (SVM) is proposed, which train SVM quickly and incrementally. In this paper, we first choose the violating KKT samples which maybe be new support vector candidates. Then for a given new-added ...
Comments