Distributed extreme learning machine optimization integrated framework system and method

Abstract

The invention belongs to the field of data processing technologies, and particularly relates to a distributed extreme learning machine optimization integrated framework system. The distributed extreme learning machine optimization integrated framework system comprises a data distributed storage module, a Stack-Autoencoder feature extraction module, a distributed calculation module and a result output module, wherein the data distributed storage module stored data in a distributed manner, and analyzes and determines an ELM hidden layer; the Stack-Autoencoder feature extraction module carries out feature learning on the data, obtains compressed input data, and performs normalization processing and feature extraction on the input data; and the distributed calculation module obtains an overall optimal weight through carrying out mapping and reduction processing on the input data. The distributed extreme learning machine optimization integrated framework system is more accurate in pattern classification of big data, solves the overfitting problem caused by too much nodes of a single-layer ELM, allows the calculation partitioned blocks of a high-dimensional matrix to be conducted in parallel, increases calculation efficiency, and saves memory resources since data does not need to be read into the memory in advance.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a distributed extreme learning machine optimization integration framework method system and method. Background technique [0002] Extreme learning machine (extreme learning machine) ELM is an easy-to-use and effective single-hidden-layer feed-forward neural network SLFNs learning algorithm. Traditional neural network learning algorithms (such as BP algorithm) need to artificially set a large number of network training parameters, and it is easy to generate local optimal solutions. The extreme learning machine only needs to set the number of hidden layer nodes of the network, and does not need to adjust the input weights of the network and the bias of the hidden elements during the algorithm execution process, and produces the only optimal solution, so it has fast learning speed and generalization The advantages of good performance. [0003] The extreme learn...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides an optimized integration system and method based on deep learning and extreme learning machine, which aims to solve the problem that the existing Basic-ELM algorithm cannot support the learning of large-scale input samples, and the features obtained thereby are not suitable for sample description. The nature is not clear enough, it is not conducive to visualization or classification, the network instability and overfitting caused by processing large data sets, and the technology based on the derived ELM algorithm that supports large data set processing causes the training set to be too large and the training time to be too long question

Images