To protect user privacy and meet law regulations, federated (machine) learning is obtaining vast interests in recent years. The key principle of training a machine model without needing to know each user’s personal raw private data. In this article, we propose secure matrix factorization framework under the setting, called FedMF. First, design user-level distributed where can be learned when on...

Tensor factorization arises in many machinelearning applications, such knowledge basemodeling and parameter estimation in latentvariable models. However, numerical meth-ods for tensor factorization have not reachedthe level of maturity of matrix factorizationmethods. In this paper, we propose a newmethod for CP tensor factorization that usesrandom projections to ...

The non-negative matrix factorization (NMF) algorithm is a classical matrix factorization and dimension reduction method in machine learning and data mining. However, in real problems, we always have to run the algorithm for several times and use the best matrix factorization result as the final output because of the random initialization of the matrix factorization. In this paper, we proposed ...

speciated samples of pm2.5 were collected at the hamshire and orange sites in golden triangle of texas by us epa (environmental protection agency) from july of 2003 to august of 2005. a total of 269 samples for the hamshire site and 293 samples for the orange site with 52 species were measured; however, 22 species were excluded because of too many below-detectionlimit data. among the 22 species...