Neural Semirings

Pedro Zuidberg Dos Martires
2021 International Workshop on Neural-Symbolic Learning and Reasoning  
From an abstract algebra perspective, reasoning tasks in symbolic artificial intelligence can be formulated in terms of commutative semiring operations. Usually, the reasoning task, e.g. probabilistic reasoning or fuzzy reasoning, is fixed upfront as part of the problem specification. In this paper we relax this assumption and render the reasoning operators (the semiring operations) subject to learning by replacing predefined operations with learnable neural networks. This unlocks the learn to reason paradigm in the semiring reasoning setting.
dblp:conf/nesy/Martires21 fatcat:jmkwhvi2b5he5lhvjx3utyxo6q