Generation of Natural Traffic Sign Images Using Domain Translation with Cycle-Consistent Generative Adversarial Networks

Spata, Dominic; Horn, Daniela; Houben, Sebastian

doi:10.1109/ivs.2019.8814090

Cited by 7 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prototypical images are also used in modern approaches, especially when dealing with small datasets. They can, for example, be the basis for creating synthetic training data (Spata et al [2019], or be employed with autoencoders for one-shot learning (Kim et al [2019].…”

Section: Related Workmentioning

confidence: 99%

Informed Pre-Training on Prior Knowledge

Rueden¹,

Houben²,

Cvejoski³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

When training data is scarce, the incorporation of additional prior knowledge can assist the learning process. While it is common to initialize neural networks with weights that have been pre-trained on other large data sets, pre-training on more concise forms of knowledge has rather been overlooked. In this paper, we propose a novel informed machine learning approach and suggest to pre-train on prior knowledge. Formal knowledge representations, e.g. graphs or equations, are first transformed into a small and condensed data set of knowledge prototypes. We show that informed pre-training on such knowledge prototypes (i) speeds up the learning processes, (ii) improves generalization capabilities in the regime where not enough training data is available, and (iii) increases model robustness. Analyzing which parts of the model are affected most by the prototypes reveals that improvements come from deeper layers that typically represent high-level features. This confirms that informed pre-training can indeed transfer semantic knowledge. This is a novel effect, which shows that knowledge-based pre-training has additional and complementary strengths to existing approaches.Preprint. Under review.

show abstract

Section: Related Workmentioning

confidence: 99%