I'm excited to announce that the Special Issue "Granular Computing for Explainable Artificial Intelligence" is now active and ready to accept submissions (deadline November 2023).
More details at https://www.springer.com/journal/12559/updates/25931758
Explainable artificial intelligence (XAI) allows domain experts to understand the reasoning of AI models and validate their outcomes. This allows employing such outcomes in real-world decision-making processes. In this regard, XAI approaches can employ information granulation approaches to aggregate the data instances hierarchically and/or semantically to provide aggregated and human-understandable explanations; represent data instances in a semantically organized manner (e.g. via clustering) to find class prototypes or counterfactuals; employ symbolic or neuro-symbolic modeling to isolate portions of neural networks that are activated by specific symbols (e.g. handwritten symbols can be recognized as groups of strokes); and obtain semantically relevant information granules (e.g. via representation learning) to be employed as concepts for building the explanations. The topics of interest include, but are not limited to:
- Intrinsically explainable granular approaches
- Information granulation approaches with explainable semantic
- Concept-based representation learning
- Natural Language Explanation
- Evaluative AI
- XAI for Fairness
- Granular Approaches to explain Latent Representations
- Clustering approaches to find prototypes, factual and counterfactual explanations
- Symbolic and neuro-symbolic modeling for XAI
- Granular model (explanation, interpretation, visualization) for decision augmentation and automation
- Application of granular XAI (healthcare, social media, multiagent systems, predictive process monitoring, etc.)
Submit your contribution at http://www.editorialmanager.com/cogn/
by selecting “SI: Granular Computing for Explainable Artificial Intelligence” for the special issue under "Additional Information."
Feel free to reach me if you have any questions.
Thank you for your contribution.
Antonio Luca Alfeo