Depending on the amount of data to process, file generation may take longer.

If it takes too long to generate, you can limit the data by, for example, reducing the range of years.


Download BibTeX


Learning Abstract Visual Reasoning via Task Decomposition: A Case Study in Raven Progressive Matrices


[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ D ] phd student | [ P ] employee

Scientific discipline (Law 2.0)

[2.3] Information and communication technology

Year of publication


Published in

International Journal of Applied Mathematics and Computer Science

Journal year: 2024 | Journal volume: vol. 34 | Journal number: no. 2

Article type

scientific article

Publication language


  • abstract visual reasoning
  • Raven Progressive Matrices
  • machine learning
  • problem decomposition

EN Learning to perform abstract reasoning often requires decomposing the task in question into intermediate subgoals that are not specified upfront, but need to be autonomously devised by the learner. In Raven Progressive Matrices (RPM), the task is to choose one of the available answers given a context, where both the context and answers are composite images featuring multiple objects in various spatial arrangements. As this high-level goal is the only guidance available, learning to solve RPMs is challenging. In this study, we propose a deep learning architecture based on the transformer blueprint which, rather than directly making the above choice, addresses the subgoal of predicting the visual properties of individual objects and their arrangements. The multidimensional predictions obtained in this way are then directly juxtaposed to choose the answer. We consider a few ways in which the model parses the visual input into tokens and several regimes of masking parts of the input in self-supervised training. In experimental assessment, the models not only outperform state-of-the-art methods but also provide interesting insights and partial explanations about the inference. The design of the method also makes it immune to biases that are known to be present in some RPM benchmarks.

Pages (from - to)

309 - 321


License type

CC BY-NC-ND (attribution - noncommercial - no derivatives)

Open Access Mode

open journal

Open Access Text Version

final published version

Ministry points / journal


Impact Factor

1,6 [List 2023]

This website uses cookies to remember the authenticated session of the user. For more information, read about Cookies and Privacy Policy.