Sebastian Bosse

Comparison of Uni- and Multimodal Interfaces for Spatial Human-Robot Interaction

2023 · Preprints.org

Multimodal user interfaces hold the promise to enable natural and intuitive means of human-machine interaction. But is the effort of using many sensors rewarded by the users - or is a single modality sufficient? For …

From attribution maps to human-understandable explanations through Concept Relevance Propagation

2023 · Nature Machine Intelligence

Abstract The field of explainable artificial intelligence (XAI) aims to bring transparency to today’s powerful but opaque deep learning models. While local XAI methods explain individual predictions in the form of attribution maps, thereby identifying …