IGANN Sparse: Bridging Sparsity and Interpretability with Non-linear Insight

1 code implementation • 17 Mar 2024 • Theodor Stoecker, Nico Hambauer, Patrick Zschech, Mathias Kraus

In this paper, we propose IGANN Sparse, a novel machine learning model from the family of generalized additive models, which promotes sparsity through a non-linear feature selection process during training.

Additive models feature selection

GAM(e) changer or not? An evaluation of interpretable machine learning models based on additive model constraints

2 code implementations • 19 Apr 2022 • Patrick Zschech, Sven Weinzierl, Nico Hambauer, Sandra Zilker, Mathias Kraus

The number of information systems (IS) studies dealing with explainable artificial intelligence (XAI) is currently exploding as the field demands more transparency about the internal decision logic of machine learning (ML) models.

Additive models Explainable artificial intelligence +2

A Light in the Dark: Deep Learning Practices for Industrial Computer Vision

1 code implementation • 6 Jan 2022 • Maximilian Harl, Marvin Herchenbach, Sven Kruschel, Nico Hambauer, Patrick Zschech, Mathias Kraus

In recent years, large pre-trained deep neural networks (DNNs) have revolutionized the field of computer vision (CV).