Learning Algorithm For Vector Quantization Research Articles

Learning algorithms for vector quantization using vertically partitioned data with IoT

The use of cloud computing systems that support AI is expanding. On the other hand, as the number of terminals and devices connected to the system increases, the decline in capacity is becoming apparent. Edge (or fog) computing system is known as one for improving a conventional cloud system. Then, how should machine learning be realized on the edge system? SMC (Secure Multiparty Computation) is known as one model to perform secure learning methods. Horizontally and vertically partitioned data are used as data structure for SMC. There have been proposed some methods of machine learning using horizontally partitioned data of SMC on the edge system. On the other hand, few studies have been done on methods using the vertically partitioned data. In this paper, fast and secure vector quantization algorithms for classification problems on vertically partitioned data with an edge system are proposed. The effectiveness is shown by numerical simulations.

Dictionary learning for VQ feature extraction in ECG beats classification

Vector quantization(VQ) can perform efficient feature extraction from electrocardiogram (ECG) with the advantages of dimensionality reduction and accuracy increase. However, the existing dictionary learning algorithms for vector quantization are sensitive to dirty data, which compromises the classification accuracy. To tackle the problem, we propose a novel dictionary learning algorithm that employs k-medoids cluster optimized by k-means++ and builds dictionaries by searching and using representative samples, which can avoid the interference of dirty data, and thus boost the classification performance of ECG systems based on vector quantization features. We apply our algorithm to vector quantization feature extraction for ECG beats classification, and compare it with popular features such as sampling point feature, fast Fourier transform feature, discrete wavelet transform feature, and with our previous beats vector quantization feature. The results show that the proposed method yields the highest accuracy and is capable of reducing the computational complexity of ECG beats classification system. The proposed dictionary learning algorithm provides more efficient encoding for ECG beats, and can improve ECG classification systems based on encoded feature.

A novel entropy-constrained competitive learning algorithm for vector quantization

In this paper, we present a novel competitive learning algorithm for the design of a variable-rate vector quantizer (VQ). The algorithm, termed entropy-constrained competitive learning (ECCL) algorithm, can achieve a near-optimal performance subject to the average rate constraint. Simulation results show that, under the same average rate, the ECCL algorithm enjoys a better performance than that of the existing competitive learning algorithms. Moreover, the ECCL algorithm outperforms the entropy-constrained vector quantizer (ECVQ) (Chou et al., IEEE Trans. Acoust. Speech Signal Process. 37 (1989) 31–42) design algorithm under the same rate constraint and initial codewords. The ECCL algorithm is also more insensitive to the selection of initial codewords as compared with the ECVQ algorithm. Therefore, the ECCL algorithm can be an effective alternative to the existing variable-rate VQ design algorithms for the applications of signal compression.

Competitive learning algorithms for vector quantization

We compare a number of training algorithms for competitive learning networks applied to the problem of vector quantization for data compression. A new competitive-learning algorithm based on the “conscience” learning method is introduced. The performance of competitive learning neural networks and traditional non-neural algorithms for vector quantization is compared. The basic properties of the algorithms are discussed and we present a number of examples that illustrate their use. The new algorithm is shown to be efficient and yields near-optimal results. This algorithm is used to design a vector quantizer for a speech database. We conclude with a discussion of continuing work.