Boolean Connectives and Deep Learning: Three Interpretations

Miguel-Angel Mendez-Lucero, Vaishak Belle

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)peer-review

Abstract / Description of output

In this paper, we motivate three approaches for integrating symbolic logic systems and deep learning methods. First, we consider whether the hidden layers of neural networks can be used to represent and reason about Boolean functions via so-called Tractable Circuits. Second, we discuss a method for encoding domain knowledge into the training and outputs of neural networks via so-called MultiplexNets. Finally, we show how we can instantiate deep learning architectures that perform exact function learning, via so-called Signal Perceptrons.
Original languageEnglish
Title of host publicationCompendium of Neurosymbolic Artificial Intelligence
EditorsPascal Hitzler, Md Kamruzzaman Sarker, Aaron Eberhart
PublisherIOS Press
ISBN (Electronic)9781643684079
ISBN (Print)9781643684062
Publication statusPublished - 2023

Publication series

NameFrontiers in Artificial Intelligence and Applications
ISSN (Print)0922-6389
ISSN (Electronic)1879-8314


Dive into the research topics of 'Boolean Connectives and Deep Learning: Three Interpretations'. Together they form a unique fingerprint.

Cite this