Visual Question Answering with Reasoning

Student: Podtikhov Artur

Supervisor: Aleksandr I. Panov

Educational Programme: Applied Mathematics and Information Science (Bachelor)

Final Grade: 10

Year of Graduation: 2020

At the moment, the task of visual question answering with reasoning is gaining popularity. Many different approaches have been proposed, but so far, the quality of existing models is incomparable with human performance. In this paper, the main existing approaches are considered and analyzed. Implemented and modified state of the art model, and also analyzed the errors of the base model, in order to highlight various trends in errors. After analyzing the errors, some additional modules are presented and implemented, the purpose of which is to fight with different types of model errors. At the end of the work, a modified model is presented, which outperforms the baseline model by 1.4%.

Full text (added May 20, 2020)

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