Learning Deep Neural Networks in a Distributed Computing Environment

Student: Almasian Sanasar

Educational Programme: Data Science and Business Analytics (Bachelor)

Final Grade: 9

Year of Graduation: 2024

This thesis examines various strategies for distributed deep learning, enhancing them through numerous optimization techniques. The focus is on decentralized synchronous distributed environments featuring various node communication patterns like fully connected all-reduce, ring, and double ring. Significant attention is given to optimizing a synchronous distributed environment capable of efficiently managing nodes with varying computational resources. Various adjustable parameters were introduced across the developed optimization techniques. Comparative results highlight the performance of the distributed environment against a locally trained model, emphasizing the impact of different parameters on the quality of the trained model.

Full text (added May 25, 2024)

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses