# EvoLearner: Learning Description Logics with Evolutionary Algorithms

arXiv: https://arxiv.org/abs/2111.04879 · HTML: https://mlsanity.com/paper/2111.04879/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 55 references that are in the map.
- First posted on arXiv 8 Nov 2021.

## Builds on (most cited first, 9 of 9)

- [A Survey Of Methods For Explaining Black Box Models](https://mlsanity.com/paper/1802.01933/) (2018)
- [Knowledge Graphs](https://mlsanity.com/paper/2003.02320/) (2020)
- [Inductive logic programming at 30: a new introduction](https://mlsanity.com/paper/2008.07912/) (2020)
- [Convolutional Complex Knowledge Graph Embeddings](https://mlsanity.com/paper/2008.03130/) (2020)
- [Convolutional Hypercomplex Embeddings for Link Prediction](https://mlsanity.com/paper/2106.15230/) (2021)
- [A Holistic Natural Language Generation Framework for the Semantic Web](https://mlsanity.com/paper/1911.01248/) (2019)
- [NABU - Multilingual Graph-based Neural RDF Verbalizer](https://mlsanity.com/paper/2009.07728/) (2020)
- [DRILL-- Deep Reinforcement Learning for Refinement Operators in ALC](https://mlsanity.com/paper/2106.15373/) (2021)
- [A Comparative Study of Distributional and Symbolic Paradigms for Relational Learning](https://mlsanity.com/paper/1806.11391/) (2018)

## Built on by (most cited first, 17 of 17)

- [Neural Class Expression Synthesis](https://mlsanity.com/paper/2111.08486/) (2021)
- [LitCQD: Multi-Hop Reasoning in Incomplete Knowledge Graphs with Numeric Literals](https://mlsanity.com/paper/2304.14742/) (2023)
- [Utilizing Description Logics for Global Explanations of Heterogeneous Graph Neural Networks](https://mlsanity.com/paper/2405.12654/) (2024)
- [Hardware-agnostic Computation for Large-scale Knowledge Graph Embeddings](https://mlsanity.com/paper/2207.08544/) (2022)
- [OntoLearner: A Modular Python Library for Ontology Learning with Large Language Models](https://mlsanity.com/paper/2607.01977/) (2026)
- [Ontolearn-A Framework for Large-scale OWL Class Expression Learning in Python](https://mlsanity.com/paper/2510.11561/) (2025)
- [Learning Permutation-Invariant Embeddings for Description Logic Concepts](https://mlsanity.com/paper/2303.01844/) (2023)
- [Reverse Engineering of Temporal Queries Mediated by LTL Ontologies](https://mlsanity.com/paper/2305.01248/) (2023)
- [SAT-Based Bounded Fitting for the Description Logic ALC](https://mlsanity.com/paper/2507.21752/) (2025)
- [Neural Reasoning for Robust Instance Retrieval in SHOIQ](https://mlsanity.com/paper/2510.20457/) (2025)
- [Can You Tell the Difference? Contrastive Explanations for ABox Entailments](https://mlsanity.com/paper/2511.11281/) (2025)
- [Explaining ELH Concept Descriptions through Counterfactual Reasoning](https://mlsanity.com/paper/2301.05109/) (2023)
- [Improving rule mining via embedding-based link prediction](https://mlsanity.com/paper/2406.10144/) (2024)
- [Discrete Diffusion-Based Model-Level Explanation of Heterogeneous GNNs with Node Features](https://mlsanity.com/paper/2508.08458/) (2025)
- [Explainable Benchmarking through the Lense of Concept Learning](https://mlsanity.com/paper/2510.20439/) (2025)
- [Semantics-Aware Caching for Concept Learning](https://mlsanity.com/paper/2603.06506/) (2026)
- [Bounded Fitting for Expressive Description Logics](https://mlsanity.com/paper/2605.07452/) (2026)

## Abstract (arXiv)

Classifying nodes in knowledge graphs is an important task, e.g., for predicting missing types of entities, predicting which molecules cause cancer, or predicting which drugs are promising treatment candidates. While black-box models often achieve high predictive performance, they are only post-hoc and locally explainable and do not allow the learned model to be easily enriched with domain knowledge. Towards this end, learning description logic concepts from positive and negative examples has been proposed. However, learning such concepts often takes a long time and state-of-the-art approaches provide limited support for literal data values, although they are crucial for many applications. In this paper, we propose EvoLearner - an evolutionary approach to learn concepts in ALCQ(D), which is the attributive language with complement (ALC) paired with qualified cardinality restrictions (Q) and data properties (D). We contribute a novel initialization method for the initial population: starting from positive examples, we perform biased random walks and translate them to description logic concepts. Moreover, we improve support for data properties by maximizing information gain when deciding where to split the data. We show that our approach significantly outperforms the state of the art on the benchmarking framework SML-Bench for structured machine learning. Our ablation study confirms that this is due to our novel initialization method and support for data properties.