Kandinsky Patterns

Kandinsky Patterns are mathematically describable, simple self-contained hence controllable test data sets for the development, validation and training of explainability in artificial intelligence.

Data and Resources

Cite this as

Heimo M¨uller, Andreas Holzinger (2024). Dataset: Kandinsky Patterns. https://doi.org/10.57702/kogbmdvf

DOI retrieved: December 16, 2024

Additional Info

Field Value
Created December 16, 2024
Last update December 16, 2024
Defined In https://doi.org/10.1016/j.artint.2021.103546
Author Heimo M¨uller
More Authors
Andreas Holzinger
Homepage https://github.com/human-centered-ai-lab/dat-kandinsky-patterns