Sparse Explanations of Neural Networks Using Pruned Layer-Wise Relevance Propagation

no code implementations • 22 Apr 2024 • Paulo Yanez Sarmiento, Simon Witzke, Nadja Klein, Bernhard Y. Renard

Our approach enforces sparsity directly by pruning the relevance propagation for the different layers.

Identifying Drivers of Predictive Uncertainty using Variance Feature Attribution

no code implementations • 12 Dec 2023 • Pascal Iversen, Simon Witzke, Katharina Baum, Bernhard Y. Renard

Identifying the drivers of uncertainty complements explanations of point predictions in recognizing potential model limitations.

Uncertainty Quantification

SimbaML: Connecting Mechanistic Models and Machine Learning with Augmented Data

1 code implementation • 8 Apr 2023 • Maximilian Kleissl, Lukas Drews, Benedict B. Heyder, Julian Zabbarov, Pascal Iversen, Simon Witzke, Bernhard Y. Renard, Katharina Baum

Training sophisticated machine learning (ML) models requires large datasets that are difficult or expensive to collect for many applications.

Benchmarking Data Augmentation +1

HiClass: a Python library for local hierarchical classification compatible with scikit-learn

3 code implementations • 13 Dec 2021 • Fábio M. Miranda, Niklas Köhnecke, Bernhard Y. Renard

HiClass is an open-source Python library for local hierarchical classification entirely compatible with scikit-learn.

Classification