Classifying Ethical AI Decisions with Explainable Prototype Based Deep Learning

Publication date

DOI

Document Type

Master Thesis

Collections

Open Access logo

License

CC-BY-NC-ND

Abstract

We propose the NorMMo architecture, that integrates neuro-fuzzy classifiers to classify the social interpretation of some set of behavioral parameters in different social situations. This architecture can facilitate the classification of norm-complying or norm-violating behaviors for a social agent in an explainable and transparent manner.

Keywords

machine ethics; social norms; machine learning; explainability; social computing; human-centered AI; responsible AI; online-learning

Citation