Paper presented at 19th International Conference on Neurosymbolic Learning and Reasoning

We had the opportunity to present our research on "Neurosymbolic Learning in Structured Probability Spaces" at the 19th Conference on Neurosymbolic Learning and Reasoning (NeSy25) at Santa Cruz, California. 

In our case study we examined the impact of neurosymbolic learning on sequence analysis in Structured Probability Spaces (SPS), comparing its effectiveness against a purely neural approach.
The results showed that NeSy system clearly surpassed their pure neural counterpart in terms of Sample Efficiency, Generalisability and Zero-Shot learning abilities, thus paving the way for more intelligible recognition systems which need less training data than most Deep Learning approaches.

For everyone who is more interested in this topic, please look at our paper which was published in Proceedings of Machine Learning Research (PMLR): https://proceedings.mlr.press/v284/fenske25a.html