Christopher M. Bishop · 2000
The Support Vector Machine (SVM) of Vapnik [9] has become widely established as one of the leading approaches to pattern recognition and machine learning. It expresses predictions in terms of a linear combination of kernel functions centred on a subset of the training data, known as support vectors. Despite its widespread success, the SVM suffers from some important limitations, one of the most significant being that it makes point predictions rather than generating predictive distributions. Recently Tipping [8] has formulated the Relevance Vector Machine (RVM), a probabilistic model whose
Cited by 0
Mohammad Reza Jabbari, Mohammad Mahdi Dorafshan, Saeid Eslamian · Elsevier eBooks · 2023
Cited by 2
Nicholas Cummins, Vidhyasaharan Sethu, Julien Epps et al. · 2015
Cited by 16
Pijush Samui, Venkata Ravibabu Mandla, Arun Krishna et al. · 2011
This article adopts Support Vector Machine (SVM) and Relevance Vector Machine (RVM) for prediction of rainfall in Vellore (India). SVM is firmly based on the theory of statistical learning theory. RVM is a probabilistic basis model. SVM and RVM use air temperature (T), sunshine, humidity and wind speed (V a) as input variables. This article uses SVM and RVM as a regression technique. Equations have been also developed for prediction of rainfall. The developed RVM gives variance of the predicted rainfall. This study shows the RVM is more robust model than the SVM.
Cited by 27
Thayananthan, A, Navaratnam, R, Stenger, B et al. · Cambridge University Engineering Department Publications Database · 2006
Cited by 0
Aaron D'Souza, Sethu Vijayakumar, Stefan Schaal · 2004
Traditional non-parametric statistical learning techniques are often computationally attractive, but lack the same generalization and model selection abilities as state-of-the-art Bayesian algorithms which, however, are usually computationally prohibitive. This paper makes several important contributions that allow Bayesian learning to scale to more complex, real-world learning scenarios. Firstly, we show that backfitting --- a traditional non-parametric, yet highly e#cient regression tool --- can be derived in a novel formulation within an expectation maximization (EM) framework and thus can finally be given a probabilistic interpretation. Secondly, we show that the general framework of sparse Bayesian learning and in particular the relevance vector machine (RVM), can be derived as a highly e#cient algorithm using a Bayesian version of backfitting at its core. As we demonstrate on several regression and classification benchmarks, Bayesian backfitting o#ers a compelling alternative to current regression methods, especially when the size and dimensionality of the data challenge computational resources.
Cited by 0
Sethu Vijayakumar, Stefan Schaal · 2004
Traditional non-parametric statistical learning techniques are often computationally attractive, but lack the same generalization and model selection abilities as state-of-the-art Bayesian algorithms which, however, are usually computationally prohibitive. This paper makes several important contributions that allow Bayesian learning to scale to more complex, real-world learning scenarios. Firstly, we show that backfitting — a traditional non-parametric, yet highly efficient regression tool — can be derived in a novel formulation within an expectation maximization (EM) framework and thus can finally be given a probabilistic interpretation. Secondly, we show that the general framework of sparse Bayesian learning and in particular the relevance vector machine (RVM), can be derived as a highly efficient algorithm using a Bayesian version of backfitting at its core. As we demonstrate on several regression and classification benchmarks, Bayesian backfitting offers a compelling alternative to current regression methods, especially when the size and dimensionality of the data challenge computational resources. 1.
Cited by 0
S. Karpagachelvi, Kodaikanal Tamilnadu India · 2014
Abstract—The ECG is one of the most effective diagnostic tools to detect cardiac diseases. It is a method to measure and record different electrical potentials of the heart. The electrical potential generated by electrical activity in cardiac tissue is measured on the surface of the human body. Current flow, in the form of ions, signals contraction of cardiac muscle fibers leading to the heart's pumping action. This ECG can be classified as normal and abnormal signals. In this paper, a thorough experimental study was conducted to show the superiority of the generalization capability of the Relevance Vector Machine (RVM) compared with Extreme Learning Machine (ELM) approach in the automatic classification of ECG beats. The generalization performance of the ELM classifier has not achieved the nearest maximum accuracy of ECG signal classsification. To achieve the maximum accuracy the RVM classifier design by searching for the best value of the parameters that tune its discriminant function, and upstream by looking for the best subset of features that feed the classifier. The experiments were conducted on the ECG data from the Massachusetts Institute of Technology–Beth Israel Hospital (MIT – BIH) arrhythmia database to classify five kinds of abnormal waveforms and normal beats. In particular, the sensitivity of the RVM classifier is tested and that is compared with ELM. Both the approaches are compared by giving raw input data and preprocessed data. The obtained results clearly confirm the superiority of the RVM approach when compared to traditional classifiers. Index Terms—Electrocardiogram (ECG) signals classification, feature detection, feature reduction, generalization capability, model selection issue, extreme learning machine (ELM), relevance vector machine (RVM). I.
Cited by 0
Shohei Tateishi, 立石, 正平, Satoshi Konishi et al. · Kyushu University Institutional Repository (QIR) (Kyushu University) · 2011
We consider the problem of variable selection in the case that explanatory variables have some groups. We proposed the extension of relevance vector machine (Tipping, 2001) for variable selection at a group level. In order to estimate a model, we derive a new estimation algorithm along with traditional relevance vector machine. Simulation results demonstrate that our methodology performs well in various situations.
Cited by 0Open access
Depeng Yang, Getao Liang, David D. Jenkins et al. · 2011
has been widely utilized in many applications, such as machine learning, image pattern recognition, and compressed sensing. However, the RVM algorithm is computationally expensive. We seek to accelerate the RVM algorithm computation for time sensi-tive applications by utilizing massively parallel accelerators such as GPUs. In this paper, the computation procedure of the RVM algorithm is fully analyzed. Recursive Cholesky decomposition, the key step in the RVM algorithm, is implemented on GPUs. The GPU performance is compared with a CPU using LAPACK and a hybrid system using the MAGMA library. Results show that our GPU implementation in both single and double precision is approximately 4 times faster than the CPU using LAPACK and faster than the hybrid MAGMA code when the matrix size is small. I.
Cited by 12
Panpan Wu, Kewen Xia, Hengyong Yu · Journal of Medical Imaging and Health Informatics · 2016
Cited by 6
Hanson Prihantoro Putro, Umi Laili Yuhana, Eko Mulyanto Yuniarno et al. · 2024
Various approaches have been proposed to detect code smells, including machine learning models. However, there are still challenges to improving detection accuracy and selecting appropriate quality metrics. This research introduces the use of Relevance Vector Machine (RVM) for code smell detection, especially Magic Value smells, which often affect source code readability and maintainability in a software development industry. This approach utilizes source code analysis using two source code representations (raw source code and token stream) and feature extraction using two text mining techniques (Bag of Words/BoW and Term Frequency-Inverse Document Frequency/TF-IDF). The experiment involves analyzing 983 records of labeled source code datasets. The results show that RVM can detect code smells with a very high level of accuracy, reaching 99.9% on some kernel configurations, also nearly perfect precision, and F1 scores. This paper contributes to introducing the use of RVM for detecting code smells in source code, explores source code representation and feature extraction techniques, and shows that RVM can achieve high detection accuracy.
Cited by 0
Arasanathan Thayananthan · 2013
The aim of this report is to detail the implementation of a sparse Bayesian Mixture of Experts (ME) [2] for solving a one-to-many regression mapping based on the relevance vector machine architecture. Our eventual goal is to evaluate the ME framework in human body and hand pose estimation from monocular view. However, this is left for future work. The application of ME is demonstrated using a toy dataset. 1 Problem Statement Given a set of input vector-output vector pairs, {z (n) , x (n) } N n=1, our aim is to learn a model of the generation process of the outputs from the inputs. At the same time, we would also like to learn an associated conditional probability distribution, p(x|z, Ω), where Ω is the set of model parameters. The learned model and the distribution can be then used to predict output x for a query input z. 1.1 Basis functions Many machine learning techniques either directly use the individual components of the input z or derive a number of non-linear kernel functions from the input. We generalise notation for both cases by defining a set of basis functions φ(z) ∈ R d derived from the input z. For example,
Cited by 0
Dongdong Kong, Yongjie Chen, Ning Li et al. · Mechanical Systems and Signal Processing · 2019
Cited by 142
Changyuan Liu, Yan Li, Xiaojun Bi · 2011
A new face recognition approach based on relevance vector machine (RVM) is presented in this paper. Firstly, the wavelet transform is applied to preprocessing face image to reduce the impact of expression change. Then, the principal component analysis (PCA) method is used to extract key features of the processed face image. Finally, the RVM classification model is adopted for identifying. In comparison with the support vector machine (SVM) method, the RVM approach performs well and can obtain more satisfactory results in terms of recognition rates and robustness.
Cited by 2
Wu Xiong, Xianhao Shen, H.-J. Li et al. · Research Square · 2021
Abstract Real-time prediction of traffic flow values in a short period of time is an importantelement in building a traffic management system. The uncertainty, complexity andnonlinearity of traffic flow data make it difficult to predict traffic flow in real time,and the accurate traffic flow prediction has been an urgent problem in the industry.Based on the research of scholars, a traffic flow prediction model based on thecorrelation vector machine method is constructed. The prediction accuracy of thecorrelation vector machine is better than that of the logistic regression and supportvector machine methods, and the correlation vector machine method has the functionof generating prediction error range for the actual traffic sequence data. Theprediction results are very satisfactory, and the prediction speed is significantlyfaster than the other two models, which meets the requirement of real-time trafficflow prediction and is suitable for real-time online prediction, and the predictionaccuracy of the used method is relatively high. The three-way comparison analysisshows that the traffic flow prediction by the correlation vector machine methodcan describe the nonlinear characteristics of traffic flow change more accurately,and the model performance and real-time performance are better. The case studyshows that the traffic flow prediction model based on the correlation vector machinecan improve the speed and accuracy of prediction, which is very suitablefor traffic flow prediction estimation with real-time requirements, and provides ascientific method for real-time traffic flow measurement.
Cited by 0Open access
Sharma, Sumeet (author) · Research Repository (Delft University of Technology) · 2023
The main purpose of a radar is to detect, recognize, and track objects of interest. When it is known that only a single target is present, the matched filter is proven to be optimal detector. However, in practice, a radar scene often consists of multiple targets. For example, in air surveillance and monitoring applications, multiple aircrafts might be in the airspace. When multiple targets are to be detected the matched filter is not guaranteed to give the best results. This can happen when a strong reflector masks the signals reflected from weak reflectors, thereby resulting in missed detections. Furthermore, when the sensor resolution is low, targets that are spaced closely together may only result in a single target actually being detected. This research explores how the Relevance Vector Machine (RVM) framework may be used to achieve a better multi-target detector than the commonly used basic matched filter approach. RVM was selected to resolve the multi-target detection problem as it estimates the target locations iteratively. In this research it was shown how the RVM framework can be used to model the fluctuation of swerling I/II targets. Additionally, the RVM algorithm was modified to incorporate a notion of statistical thresholding. Simulations show that using RVM the false alarm rate can be reduced and target locations can be more accurately recovered compared to other existing methods in case of multiple swerling I/II fluctuating targets. Furthermore, the proposed approach is shown to have a much lower convergence time compared to a similar expectation-maximization based method, namely Enhanced Sparse Bayesian Learning.
Cited by 0Open access
Daniela Marconi, Valter Gattei, Renato Campanini · 2015
Proposed analytical objective In microarray data analysis most of the literature concerns the identification of significant differences in expression profiles between two or more classes. Both supervised and unsupervised algorithms are designed to distinguish, with a certain level of confidence, one class from another. However, looking at microarray data in a prognostic and diagnostic clinical framework, not only differences could have a crucial role. In some cases similarities can give useful and, sometimes even more, important information. The introduction of prognosticators based on classification algorithms, applied to disease microarray dataset, could be the next future (Vant’veer’s group is one of the pioneers in this field [1]). But one of the limitation is that at moment the response of the classificator is just a “hard ” binary decision. Instead it could be useful to obtain an estimate of the posterior probability of new sample’s membership to a class, i.e. to say how much the new sample is “similar ” to the selected class samples. Another way to interpret a measure of similarities is a three class problem. The goal, given three classes, could be to establish, with a certain level of confidence, if the third one is similar to the first or the second one.
Cited by 0
Yongfu He, Shaojun Wang, Yu Peng et al. · 2014
Relevance Vector Machine (RVM) with the uncertainty expressing ability has spawned broad applications in Prognostic and Health Management (PHM). However computationally intensive intrinsic nature of RVM greatly limits its usage. This paper presents a software and hardware co-design approach based on HMPSoC technology, which efficiently exploited sequential and parallel nature of RVM. Multi-channel and pipelined hardware architecture for the acceleration of kernel formulation and intermediate values calculation is proposed. The hardware that wrapped with AXI-Stream interface is integrated into HMPSoC as an acceleration engine. We implement the design on an on-board PHM prototype platform with a Xilinx Zynq XC7Z020 AP SoC. The experiment results show 5.3× and 46.8× speed up in terms of the time cost than the RVM running on PC with a Xeon 5620 processor and ARM Cortex A9 processor. The energy consumption is reduced by 153.0× and 37.3×, respectively.
Cited by 3
Ruchun Mo, Libo Chen, Zhiquan Xing et al. · Structures · 2023
Cited by 18