Parallel programming paradigms for decision making system in the presence of uncertainty in data

This article analyzes the possibilities of using parallel computing approaches in machine learning algorithms. The article describes methods such as data parallelism and model parallelism and their effectiveness based on examples. Parallel computing technologies are considered a key factor in the development of modern machine learning architectures. The importance of parallelism in training models working with large volumes of data, in particular, its role in speeding up calculation processes at the training stage, has been demonstrated. The efficiency of using parallel computing is demonstrated by a computational experiment.

Ҳали таржима қилинмаган