Learning Finite Linear Temporal Logic Specifications with a Specialized Neural Operator

Abstract

Finite linear temporal logic (LTL_f) is a powerful formal representation for modeling temporal sequences. We address the problem of learning a compact LTL_f formula from labeled traces of system behavior. We propose a novel neural network operator and evaluate the resulting architecture, NeuralLTL_f. Our approach includes a specialized recurrent filter, designed to subsume LTL_f temporal operators, to learn a highly accurate classifier for traces. Then, it discretizes the activations and extracts the truth table represented by the learned weights. This truth table is converted to symbolic form and returned as the learned formula. Experiments on randomly generated LTL_f formulas show NeuralLTL_f scales to larger formula sizes than existing approaches and maintains high accuracy even in the presence of noise.

Tasks

Reproductions