Integrating Intrinsic and Extrinsic Explainability: The Relevance of\n Understanding Neural Networks for Human-Robot Interaction

Abstract

Explainable artificial intelligence (XAI) can help foster trust in and\nacceptance of intelligent and autonomous systems. Moreover, understanding the\nmotivation for an agent's behavior results in better and more successful\ncollaborations between robots and humans. However, not only can humans benefit\nfrom a robot's explanation but the robot itself can also benefit from\nexplanations given to him. Currently, most attention is paid to explaining deep\nneural networks and black-box models. However, a lot of these approaches are\nnot applicable to humanoid robots. Therefore, in this position paper, current\nproblems with adapting XAI methods to explainable neurorobotics are described.\nFurthermore, NICO, an open-source humanoid robot platform, is introduced and\nhow the interaction of intrinsic explanations by the robot itself and extrinsic\nexplanations provided by the environment enable efficient robotic behavior.\n