Artificial Neural Networks that Learn to Satisfy Logic Constraints

TL;DR

This paper introduces neural architectures capable of encoding and solving logic-constrained problems through unsupervised learning, effectively integrating symbolic knowledge with neural networks for structured reasoning tasks.

Contribution

The authors present two neural models that encode relational knowledge and logic constraints, enabling unsupervised learning to solve combinatorial problems without labeled answers.

Findings

Networks learn to satisfy logic constraints through unsupervised training.

The approach improves problem-solving speed over iterations.

Potential for generalization to various structured reasoning tasks.

Abstract

Logic-based problems such as planning, theorem proving, or puzzles, typically involve combinatoric search and structured knowledge representation. Artificial neural networks are very successful statistical learners, however, for many years, they have been criticized for their weaknesses in representing and in processing complex structured knowledge which is crucial for combinatoric search and symbol manipulation. Two neural architectures are presented, which can encode structured relational knowledge in neural activation, and store bounded First Order Logic constraints in connection weights. Both architectures learn to search for a solution that satisfies the constraints. Learning is done by unsupervised practicing on problem instances from the same domain, in a way that improves the network-solving speed. No teacher exists to provide answers for the problem instances of the training and test sets. However, the domain constraints are provided as prior knowledge to a loss function that measures the degree of constraint violations. Iterations of activation calculation and learning are executed until a solution that maximally satisfies the constraints emerges on the output units. As a test case, block-world planning problems are used to train networks that learn to plan in that domain, but the techniques proposed could be used more generally as in integrating prior symbolic knowledge with statistical learning