Integrating symbolic and sub-symbolic approaches to make intelligent systems understandable and explainable is at the core of new fields such as neuro-symbolic computing. Neural-symbolic computing aims to integrate two fundamental cognitive abilities: learning from the environment and reasoning from what has been learned. Neural-symbolic computing reconciles the advantages of robust learning in neural networks and reasoning and interpretability of symbolic representation, enabling explainability via symbolic representations for network models.

• Probabilistic methods for explainability with Neural Networks
• Enabling explanations with Computing Logic Programs with Neural Networks
• Learning Logic Programs for explainable AI systems
• Enabling explanations with Neurosymbolic Logic Programming
• Enabling explanations with Nonmonotonic Reasoning and Learning
• Machine Learning for Theorem Proving
• Abductive Logic Programming and Learning for XAI
• Learning Answer Set Programs for XAI
• Interpretability and Explainability via symbolic knowledge extraction and injection from/to opaque models
• Enabling explanations with Inductive logic programming
• Explainable planning