Our paper on ULLER, a general-purpose NeSy language, is accepted as a spotlight at NeSy 2024.

Created in September 09, 2024

2024

Our paper on ULLER, a general-purpose NeSy language, is accepted as a spotlight oral presentation at NeSy 2024. We are also hosting a tutorial and feedback discussions on an early version of the library.

ULLER is a Python library we are developping with Emile Van Krieken, Eleonora Giunchiglia and Robin Manhaeve to simplify working with Informed Nesy AI. NeSy has been gaining attention recently but remains hard to approach/experiment with: there is a myriad of methods, each with their own knowledge representation, making it difficult to experiment and benchmark with NeSy. Our library aims to solve this. 🚀

ULLER provides:

  • A shared language for expressing problem knowledge
  • Integration with various Nesy methods
  • Easy incorporation into ML workflows

Our paper focuses on the design ofthe language and its semantics. We presented it at NeSy 2024 (spotlight) and also hosted a tutorial session and feedback discussions around a prototype version of the library.

Tutorial and discussions session on ULLER. Credits to Jaron Maene.