Online Asynchronous Distributed Regression
This work addresses the challenge of handling massive datasets efficiently for machine learning practitioners, though it appears incremental as it builds on existing gradient-type optimization models.
The paper tackled the problem of nonparametric online regression in distributed computing environments by proposing a consensus-based asynchronous distributed approach, achieving excellent performance with up to 28 parallel processors in terms of computation time and prediction accuracy.
Distributed computing offers a high degree of flexibility to accommodate modern learning constraints and the ever increasing size of datasets involved in massive data issues. Drawing inspiration from the theory of distributed computation models developed in the context of gradient-type optimization algorithms, we present a consensus-based asynchronous distributed approach for nonparametric online regression and analyze some of its asymptotic properties. Substantial numerical evidence involving up to 28 parallel processors is provided on synthetic datasets to assess the excellent performance of our method, both in terms of computation time and prediction accuracy.