23,948 Hits in 6.9 sec

Online Learning with Regularized Kernel for One-class Classification [article]

Chandan Gautam, Aruna Tiwari, Sundaram Suresh, Kapil Ahuja
2018 arXiv   pre-print
The baseline kernel hyperplane model considers whole data in a single chunk with regularized ELM approach for offline learning in case of one-class classification (OCC).  ...  This paper presents an online learning with regularized kernel based one-class extreme learning machine (ELM) classifier and is referred as online RK-OC-ELM.  ...  ACKNOWLEDGMENT This research was supported by Department of Electronics and Information Technology (DeITY, Govt. of India) under Visvesvaraya PHD scheme for electronics & IT.  ... 
arXiv:1701.04508v2 fatcat:e327bwlstfcvpekl6qragnyp7y

Online Multikernel Learning Based on a Triple-Norm Regularizer for Semantic Image Classification

Shuangping Huang, Lianwen Jin, Yunyu Li
2015 Mathematical Problems in Engineering  
Currently image classifiers based on multikernel learning (MKL) mostly use batch approach, which is slow and difficult to scale up for large datasets.  ...  In the meantime, standard MKL model neglects the correlations among examples associated with a specific kernel, which makes it infeasible to adjust the kernel combination coefficients.  ...  Triple-Norm Regularizer. According to Section 2.1, each component of associated with a specific class and kernel, that is, , ( = 1, . . . , , = 1, . . . , ), is a coefficient vector.  ... 
doi:10.1155/2015/346496 fatcat:xu3ternfzbe5hpyqx65m4jbkam

An Online Stochastic Kernel Machine for Robust Signal Classification [article]

Raghu G. Raj
2019 arXiv   pre-print
We present a novel variation of online kernel machines in which we exploit a consensus based optimization mechanism to guide the evolution of decision functions drawn from a reproducing kernel Hilbert  ...  Incorporating Stochastic Structure Kernel machines are based on the following regularized cost function for signal classification: [ ; ] = 1 ∑ ( ( ), ) =1 + 2 ‖ ‖ 2 (5) where, ‖ ‖ = 〈 , 〉 1/2 , is a loss  ...  In section 4 we apply the osKM algorithm to the problem of online signal classification with input and label noise and compare the performance to a traditional online learning counterpart.  ... 
arXiv:1905.07686v2 fatcat:ctg7radssndhjhswpjnn2yutye

Indoor Localization via Discriminatively Regularized Least Square Classification

Robin Wentao Ouyang, Albert Kai-Sun Wong, Kam Tim Woo
2011 International Journal of Wireless Information Networks  
Moreover, we address the missing value problem, utilize clustering to reduce the training and online complexity, and introduce kernel alignment for fast kernel parameter tuning.  ...  Experimental results show that, compared with other methods, the kernel DRLSC-based algorithm achieves superior performance for indoor localization when only a small fraction of the data samples are used  ...  Online Prediction After solving for W and b based on the training data, we need to perform the online prediction.  ... 
doi:10.1007/s10776-011-0133-5 fatcat:sxjtqhehjjcvtjy57rtm5hyhdm

Multi-view Multi-class Classification for Identification of Pathogenic Bacterial Strains [chapter]

Evgeni Tsivtsivadze, Tom Heskes, Armand Paauw
2013 Lecture Notes in Computer Science  
We propose multi-class classification method that is particularly suitable for multi-view learning setting.  ...  methods on publicly available datasets.  ...  It also allows efficient construction of different views for the co-regularization problem (see online supplementary material) and leads to notable speed up when dealing with large-scale learning tasks  ... 
doi:10.1007/978-3-642-38067-9_6 fatcat:y2vwkfrjobcvxjliiraozu76ky

Online learning with minority class resampling

Michael J. Pekala, Ashley J. Llorens
2011 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
We show empirically that the addition of caching and minority class oversampling to online learners improves the g-means performance under these conditions by compensating for class imbalance.  ...  This paper considers using online binary classification for target detection where the goal is to identify signals of interest within a sequence of received signals generated by a shifting background.  ...  ACKNOWLEDGEMENTS The authors would like to thank I-Jeng Wang and the reviewers for their helpful comments and suggestions.  ... 
doi:10.1109/icassp.2011.5946929 dblp:conf/icassp/PekalaL11 fatcat:67kuki32vnfsrnghi73llaqolq

Information-theoretic metric learning

Jason V. Davis, Brian Kulis, Prateek Jain, Suvrit Sra, Inderjit S. Dhillon
2007 Proceedings of the 24th international conference on Machine learning - ICML '07  
Finally, we evaluate our method on a recent error reporting system for software called Clarify, in the context of metric learning for nearest neighbor classification, as well as on standard data sets.  ...  We also present an online version and derive regret bounds for the resulting algorithm.  ...  Online Metric Learning In this section, we describe an online algorithm for metric learning and prove bounds on the total loss when using LogDet divergence as the regularizer.  ... 
doi:10.1145/1273496.1273523 dblp:conf/icml/DavisKJSD07 fatcat:nefliz47jreszkf6flomazin7e

Generating feature spaces for linear algorithms with regularized sparse kernel slow feature analysis

Wendelin Böhmer, Steffen Grünewälder, Hannes Nickisch, Klaus Obermayer
2012 Machine Learning  
We make use of the kernel trick in combination with sparsification to develop a kernelized SFA algorithm which provides a powerful function class for large data sets.  ...  Non-linear optimization of this unsupervised learning method generates an orthogonal basis on the unknown latent space for a given time series.  ...  learning), the German Federal Ministry of Education and Research (grant 01GQ0850) and EPSRC grant #EP/H017402/1 (CARDyAL).  ... 
doi:10.1007/s10994-012-5300-0 fatcat:taybe3sdebggrnjp4xymxhfcga

Online Learning: A Comprehensive Survey [article]

Steven C.H. Hoi, Doyen Sahoo, Jing Lu, Peilin Zhao
2018 arXiv   pre-print
where full feedback information is always available, (ii) online learning with limited feedback, and (iii) unsupervised online learning where there is no feedback available.  ...  Online learning represents an important family of machine learning algorithms, in which a learner attempts to resolve an online prediction (or any type of decision-making) task by learning a model/hypothesis  ...  As a consequence, solving a regular online kernel classification task can be turned into a linear online classification task on the new feature space derived from the kernel approximation.  ... 
arXiv:1802.02871v2 fatcat:mqorsb4gknhfhjfb4jcsvbrtwm

Scalable Nonlinear AUC Maximization Methods [article]

Majdi Khalid, Indrakshi Ray, Hamidreza Chitsaz
2019 arXiv   pre-print
It has been widely used to evaluate classification performance on heavily imbalanced data.  ...  However, the high training complexity renders the kernelized AUC machines infeasible for large-scale data.  ...  However, online methods result in inferior classification accuracy compared to batch learning algorithms.  ... 
arXiv:1710.00760v4 fatcat:dqpgvvabyjcczbk4373ovsctjq

Online Multiple Kernel Classification

Steven C. H. Hoi, Rong Jin, Peilin Zhao, Tianbao Yang
2012 Machine Learning  
The proposed algorithms are based on the fusion of two online learning algorithms, i.e., the Perceptron algorithm that learns a classifier for a given kernel, and the Hedge algorithm that combines classifiers  ...  As an attempt to fill the gap, we address a new research problem, termed Online Multiple Kernel Classification (OMKC), which learns a kernel-based prediction function by selecting a subset of predefined  ...  Proposed framework for online classification with multiple kernels We introduce the problem setting and regular Multiple Kernel Learning (MKL), and then present the proposed framework of online multiple  ... 
doi:10.1007/s10994-012-5319-2 fatcat:hsxs6nmi6fba5cajfrsww6sg2a

Online learning in biometrics: A case study in face classifier update

Richa Singh, Mayank Vatsa, Arun Ross, Afzel Noore
2009 2009 IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems  
Specifically, the algorithm employs online learning technique in a 2ν-Granular Soft Support Vector Machine for rapidly training and updating face recognition systems.  ...  To account for the variations in data distribution caused by these new enrollments, biometric systems require regular re-training which usually results in a very large computational overhead.  ...  FORMULATION OF ONLINE LEARNING FOR 2ν-GSSVM In general, SVM is used for a two-class classification problem.  ... 
doi:10.1109/btas.2009.5339071 fatcat:puwpovl4v5g47nh74x53psy2im

Comparison of Supervised Classification Models on Textual Data

Bi-Min Hsu
2020 Mathematics  
Further exploration on the use of different SVM kernels was performed, demonstrating the advantage of using linear kernels over polynomial, sigmoid, and radial basis function kernels for text classification  ...  With the growing number of textual documents and datasets generated through social media and news articles, an increasing number of machine learning methods are required for accurate textual classification  ...  Text classification is important for categorizing online content, such as topic detection of online news articles and spam detection in emails.  ... 
doi:10.3390/math8050851 fatcat:b6ebhrr6brckjm2jggxg7wrm6q

OM-2: An online multi-class Multi-Kernel Learning algorithm Luo Jie

Francesco Orabona, Marco Fornoni, Barbara Caputo, Nicolo Cesa-Bianchi
2010 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops  
In this work we introduce a theoretically motivated and efficient online learning algorithm for the Multi Kernel Learning (MKL) problem.  ...  For this algorithm we prove a theoretical bound on the number of multiclass mistakes made on any arbitrary data sequence.  ...  Acknowledgments The Caltech-101 kernel matrices were kindly provided by Peter Gehler. LJ, MF and BC were supported by the EU project DIRAC IST-027787.  ... 
doi:10.1109/cvprw.2010.5543766 dblp:conf/cvpr/OrabonaFCC10 fatcat:em4g32warbbyjorr7kkznnrotu

Online spatio-temporal Gaussian process experts with application to tactile classification

Harold Soh, Yanyu Su, Yiannis Demiris
2012 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems  
In this work, we are primarily concerned with robotic systems that learn online and continuously from multivariate data-streams.  ...  Our first contribution is a new recursive kernel, which we have integrated into a sparse Gaussian Process to yield the Spatio-Temporal Online Recursive Kernel Gaussian Process (STORK-GP).  ...  ACKNOWLEDGEMENT The authors thank members of the Personal Robotics Laboratory at Imperial College London, particularly Yan Wu for his assistance in the iCub experiments.  ... 
doi:10.1109/iros.2012.6385992 dblp:conf/iros/SohSD12 fatcat:tgivaoe6kvfz7pfhpyh4nffafa
« Previous Showing results 1 — 15 out of 23,948 results