Solving Abstract Reasoning Tasks with Grammatical Evolution
The Abstraction and Reasoning Corpus (ARC) comprising image-based logical reasoning tasks is intended to serve as a benchmark for measuring intelligence. Solving these tasks is very difficult for off-the-shelf ML methods due to heterogeneous semantics and low amount of training data. We here present our approach, which solves tasks via grammatical evolution on a domain-specific language for image transformations. With this approach, we successfully participated in an online challenge, scoring among the top 4% out of 900 participants.
- Published in:
Proceedings of the LWDA 2020 Lernen. Wissen. Daten. Analysen. (LWDA) - Type:
Inproceedings - Authors:
R. Fischer, M. Jakobs, S. Mücke, K. Morik - Year:
2020
Citation information
R. Fischer, M. Jakobs, S. Mücke, K. Morik: Solving Abstract Reasoning Tasks with Grammatical Evolution, Lernen. Wissen. Daten. Analysen. (LWDA), Proceedings of the LWDA 2020, 2020, https://www.researchgate.net/publication/348408303_Solving_Abstract_Reasoning_Tasks_with_Grammatical_Evolution, Fischer.etal.2020a,
@Inproceedings{Fischer.etal.2020a,
author={R. Fischer, M. Jakobs, S. Mücke, K. Morik},
title={Solving Abstract Reasoning Tasks with Grammatical Evolution},
booktitle={Lernen. Wissen. Daten. Analysen. (LWDA)},
journal={Proceedings of the LWDA 2020},
url={https://www.researchgate.net/publication/348408303_Solving_Abstract_Reasoning_Tasks_with_Grammatical_Evolution},
year={2020},
abstract={The Abstraction and Reasoning Corpus (ARC) comprising image-based logical reasoning tasks is intended to serve as a benchmark for measuring intelligence. Solving these tasks is very difficult for off-the-shelf ML methods due to heterogeneous semantics and low amount of training data. We here present our approach, which solves tasks via grammatical evolution on a domain-specific language for image...}}