Making Logic Learnable With Neural Networks

Tobias Brudermueller, Dennis L. Shung, Adrian J. Stanley +2

While neural networks are good at learning unspecified functions from training samples, they cannot be directly implemented in hardware and are often not interpretable or formally…