A novel software module (conflicting_bundle.py) to detect problematic layers in deep neural networks is introduced.
•
conflicting_bundle.py works for arbitrary layer types.
•
conflicting_bundle.py can be used to precisely analyze and improve neural network as demonstrated in related work.
•
conflicting_bundle.py will speed up (otherwise exhaustive) grid searches.
•
conflicting_bundle.py already demonstrated its success in a novel AutoML method.
Abstract
Designing neural network architectures is a challenging task and knowing which specific layers of a neural network must be adapted to improve the performance is almost a mystery. In this paper, we introduce the conflicting_bundle.py module to identify layers that decrease the accuracy of trained networks. Therefore, this software-module helps machine-learning researchers and engineers to precisely analyze and improve neural network architectures. The same software-module can also be used to automatically create improved neural network architectures.
