nonnegativematrixfactorization (NMF) has been successfully applied in several data mining tasks. Recently, there is an increasing interest in the acceleration of NMF, due to its high cost on large matrices. On the o...
详细信息
nonnegativematrixfactorization (NMF) has been successfully applied in several data mining tasks. Recently, there is an increasing interest in the acceleration of NMF, due to its high cost on large matrices. On the other hand, the privacy issue of NMF over federated data is worthy of attention, since NMF is prevalently applied in image and text analysis which may involve leveraging privacy data (e.g, medical image and record) across several parties (e.g., hospitals). In this paper, we study the acceleration and security problems of distributed NMF. First, we propose a distributed sketched alternating nonnegative least squares(DSANLS) framework for NMF, which utilizes a matrix sketching technique to reduce the size of nonnegative least squares subproblems with a convergence guarantee. For the second problem, we show that DSANLS with modification can be adapted to the security setting, but only for one or limited iterations. Consequently, we propose four efficient distributed NMF methods in both synchronous and asynchronous settings with a security guarantee. We conduct extensive experiments on several real datasets to show the superiority of our proposed methods. The implementation of our methods is available at https://***/qianyuqiu79/DSANLS.
nonnegativematrixfactorization (NMF) is a commonlyused unsupervised learning method for extracting parts-based features and dimensionality reduction from nonnegative data. Many computational algorithms exist for upd...
详细信息
ISBN:
(纸本)9783319912622;9783319912615
nonnegativematrixfactorization (NMF) is a commonlyused unsupervised learning method for extracting parts-based features and dimensionality reduction from nonnegative data. Many computational algorithms exist for updating the latent nonnegative factors in NMF. In this study, we propose an extension of the Hierarchical Alternating Least Squares (HALS) algorithm to a distributed version using the state-of-the-art framework - Apache Spark. Spark gains its popularity among other distributed computational frameworks because of its in-memory approach which works much faster than well-known Apache Hadoop. The scalability and efficiency of the proposed algorithm is confirmed in the numerical experiments, performed on real data as well as synthetic ones.
nonnegativematrixfactorization (NMF) is a commonly used method in machine learning and data analysis for feature extraction and dimensionality reduction of nonnegative data. Recently, we observe its increasing popul...
详细信息
ISBN:
(数字)9783319654829
ISBN:
(纸本)9783319654829;9783319654812
nonnegativematrixfactorization (NMF) is a commonly used method in machine learning and data analysis for feature extraction and dimensionality reduction of nonnegative data. Recently, we observe its increasing popularity in processing massive data, and advances in developing various distributed algorithms for NMF. In the paper, we propose a computational strategy for implementation of the Hierarchical Alternating Least Squares (HALS) algorithm using the MapReduce programming paradigm. Due to this approach, the scalable HALS NMF, which can be implemented on parallel and distributed computer architectures, is obtained. The scalability and efficiency of the proposed algorithm is confirmed in the numerical experiments, performed on largescale synthetic and recommendation system datasets.
nonnegativematrixfactorization and its multilinear extension known as nonnegative tensor factorization are commonly used methods in machine learning and data analysis for feature extraction and dimensionality reduct...
详细信息
nonnegativematrixfactorization and its multilinear extension known as nonnegative tensor factorization are commonly used methods in machine learning and data analysis for feature extraction and dimensionality reduction for nonnegative high-dimensional data. Dimensionality reduction for massive amounts of data usually involves distributed computation across multi-node computer architectures. In this study, we propose various computational strategies for parallel and distributed computation of the latent factors in both factorization models, all of which are based on partitioning the computational tasks according to the MapReduce paradigm. We extend the previously reported distributed hierarchical alternating least squares algorithm to the multi-way array factorization model, where we assume that the observed multi-way data can be partitioned into chunks along one mode. Moreover, we propose a new geometry-based distributed computational strategy for solving nonnegativematrixfactorization problems. Numerical experiments performed using various large-scale data sets demonstrated that these algorithms are efficient and robust to noisy data.
暂无评论