Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Batch-wise Regularization of Deep Neural Networks for Interpretability

: Burkart, Nadia; Faller, Philipp M.; Peinsipp, Elisabeth; Huber, Marco


Institute of Electrical and Electronics Engineers -IEEE-; IEEE Robotics and Automation Society; Informationstechnische Gesellschaft -ITG-; Verband der Elektrotechnik, Elektronik, Informationstechnik -VDE-:
IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, MFI 2020 : 14-16 September 2020, virtuell, Karlsruhe, Germany
Piscataway, NJ: IEEE, 2020
ISBN: 978-1-7281-6422-9
ISBN: 978-1-7281-6421-2
ISBN: 978-1-7281-6423-6
International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI) <2020, Online>
Conference Paper
Fraunhofer IPA ()
Fraunhofer IOSB ()
Explainable Artificial Intelligence (XAI); Künstliche Intelligenz; maschinelles Lernen; neuronales Netz

Fast progress in the field of Machine Learning and Deep Learning strongly influences the research in many application domains like autonomous driving or health care. In this paper, we propose a batch-wise regularization technique to enhance the interpretability for deep neural networks (NN) by means of a global surrogate rule list. For this purpose, we introduce a novel regularization approach that yields a differentiable penalty term. Compared to other regularization approaches, our approach avoids repeated creating of surrogate models during training of the NN. The experiments show that the proposed approach has a high fidelity to the main model and also results in interpretable and more accurate models compared to some of the baselines.