Learning from data streams using kernel least-mean-square with multiple kernel-sizes and adaptive step-size

García-Vega, Sergio; Zeng, Xiao‐Jun; Keane, John

doi:10.1016/j.neucom.2019.01.055

Cited by 11 publications

(8 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, trial-and-error or cross-validation experiments are often required, which can be computationally demanding. To reduce the cost, some research work has been conducted, including Maximum Corr-entropy Criterion [7], Bayesian averaging rule [8], and stochastic gradient method [9], to optimize the kernel size during the training process.…”

Section: Kernel-based Learning Algorithmsmentioning

confidence: 99%

“…In this section, the proposed AKSL algorithm is compared with five conventional sparse training algorithms, namely CLS-SVM [6], SSS [10], SLQ [11], CCS [22], and VSS [9]. We have briefly introduced the algorithm of CLS-SVM, SSS, and SLQ in Section 2.…”

Section: Comparison With Other Training Algorithmsmentioning

confidence: 99%

“…On the other hand, the VSS method presented in [9] is to solve a least-mean-square based problem with adaptive kernel-sizes. The major difference of VSS, compared to others, is that VSS does not consider the solution sparsity, but only optimizes the kernel size.…”

Section: Comparison With Other Training Algorithmsmentioning

confidence: 99%

See 2 more Smart Citations

Sparse Least Squares Support Vector Machine With Adaptive Kernel Parameters

Yang¹,

Yang²,

Ma³

2020

IJCIS

View full text Add to dashboard Cite

In this paper, we propose an efficient Least Squares Support Vector Machine (LS-SVM) training algorithm, which incorporates sparse representation and dictionary learning. First, we formalize the LS-SVM training as a sparse representation process. Second, kernel parameters are adjusted by optimizing their average coherence. As such, the proposed algorithm addresses the training problem via generating the sparse solution and optimizing kernel parameters simultaneously. Experimental results demonstrate that the proposed algorithm is capable of achieving competitive performance compared to state-of-the-art approaches.

show abstract

Section: Kernel-based Learning Algorithmsmentioning

confidence: 99%

Section: Comparison With Other Training Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

Sparse Least Squares Support Vector Machine With Adaptive Kernel Parameters

Yang¹,

Yang²,

Ma³

2020

IJCIS

View full text Add to dashboard Cite

show abstract

“…e existing literature focuses on the multiplekernel learning method and solves different issues such as kernel size and step size. We follow the same line of thought and take the existing methods [17,19,20] as the foundation of the proposed work to propose a KAF-based approach for close-price prediction.…”

Section: Introductionmentioning

confidence: 99%

Multivariate and Online Prediction of Closing Price Using Kernel Adaptive Filtering

Mishra

Ahmed

Mishra

et al. 2021

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

This paper proposes a multivariate and online prediction of stock prices via the paradigm of kernel adaptive filtering (KAF). The prediction of stock prices in traditional classification and regression problems needs independent and batch-oriented nature of training. In this article, we challenge this existing notion of the literature and propose an online kernel adaptive filtering-based approach to predict stock prices. We experiment with ten different KAF algorithms to analyze stocks’ performance and show the efficacy of the work presented here. In addition to this, and in contrast to the current literature, we look at granular level data. The experiments are performed with quotes gathered at the window of one minute, five minutes, ten minutes, fifteen minutes, twenty minutes, thirty minutes, one hour, and one day. These time windows represent some of the common windows frequently used by traders. The proposed framework is tested on 50 different stocks making up the Indian stock index: Nifty-50. The experimental results show that online learning and KAF is not only a good option, but practically speaking, they can be deployed in high-frequency trading as well.

show abstract

“…Two well-known KAFs is the kernel least mean square (KLMS) algorithm [28] and the kernel recursive least-squares (KRLS) algorithm [29]. The KAF method has been widely applied in the field of system identification, noise elimination and time series prediction [30]- [33], while the applications in energy systems are still few and far between. Since KAF is favorite for its robustness, low complexity and high approximation capability, it is worthwhile to introduce KAF in the prognostics and health management of energy system.…”

Section: Introductionmentioning

confidence: 99%

A Robust Hybrid Filtering Method for Accurate Battery Remaining Useful Life Prediction

Peng

Gao

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Accurate remaining useful life (RUL) prediction under the noisy environment is a big challenge for the health management of modern industrial systems since the extraction of the accurate data structure from heavily corrupted data is difficult. In recent years, the kernel adaptive filter (KAF) has been widely adopted to solve the robust regression problem due to its low-complexity and high-approximation capability and robustness while the applications in battery RUL prediction are still few and far between. Thus, this paper is concerned with long-term RUL prediction using the KAF method. At first, concretely speaking, a robust KAF algorithm is derived based on the double-Gaussian-mixture (DGM) cost function, which is used to learn the capacity degradation mechanism from contaminated capacity data and so as to build the long-term prediction model. Second, a robust unscented Kalman filter (UKF) algorithm employing the DGM-based cost function is developed, which is then combined with the KAF-based prediction model to realize a more accurate and reliable prediction. Under the hybrid prognostic framework, the proposed UKF algorithm is applied to filter the noisy observations. When the observation data are inaccessible, the predicted data from the off-line trained KAF-based prediction model are adopted as the approximated value of the real observations for the UKF algorithm to optimize the prediction results and to provide the uncertainty representation. The experimental results reveal that the proposed method has great robustness when the measurements contain noise and large outliers, which makes it possible to get satisfactory prediction performance without preprocessing the data manually.INDEX TERMS Kernel adaptive filter, battery, unscented kalman filter, remaining useful life (RUL), robustness.

show abstract

Learning from data streams using kernel least-mean-square with multiple kernel-sizes and adaptive step-size

Cited by 11 publications

References 38 publications

Sparse Least Squares Support Vector Machine With Adaptive Kernel Parameters

Sparse Least Squares Support Vector Machine With Adaptive Kernel Parameters

Multivariate and Online Prediction of Closing Price Using Kernel Adaptive Filtering

A Robust Hybrid Filtering Method for Accurate Battery Remaining Useful Life Prediction

Contact Info

Product

Resources

About