Application of Transformers to Time Series Classification

Student: Avilov Maksim

Faculty: Faculty of Informatics, Mathematics, and Computer Science (HSE Nizhny Novgorod)

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Year of Graduation: 2024

Transformers have been widely used recently in solving various kinds of tasks due to their unique architecture with the attention mechanism. The use of transformers for time series analysis is widely studied, as they show high efficiency due to their ability to learn on large amounts of data and efficiently process long sequences. In this paper, a review of time series classification methods is carried out and the transformer's own architecture is tested, supplemented by the Bag-Of-SFA-Symbols mechanism. In the practical part, a comparative analysis of this approach with other popular architectures was carried out on various datasets containing ECG.

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