Weighted Graph-Based Signal Temporal Logic Inference Using Neural Networks
This work addresses the need for interpretable and formally analyzable spatial-temporal knowledge extraction, which is useful in applications such as healthcare and weather prediction, though it appears incremental as it builds on existing wGSTL frameworks.
The paper tackles the problem of extracting human-interpretable spatial-temporal knowledge from data by proposing a method that trains neural networks to learn weighted graph-based signal temporal logic (wGSTL) formulas, with results showing classification accuracy comparable to baseline methods like K-nearest neighbors and decision trees on COVID-19 and rain prediction datasets.
Extracting spatial-temporal knowledge from data is useful in many applications. It is important that the obtained knowledge is human-interpretable and amenable to formal analysis. In this paper, we propose a method that trains neural networks to learn spatial-temporal properties in the form of weighted graph-based signal temporal logic (wGSTL) formulas. For learning wGSTL formulas, we introduce a flexible wGSTL formula structure in which the user's preference can be applied in the inferred wGSTL formulas. In the proposed framework, each neuron of the neural networks corresponds to a subformula in a flexible wGSTL formula structure. We initially train a neural network to learn the wGSTL operators and then train a second neural network to learn the parameters in a flexible wGSTL formula structure. We use a COVID-19 dataset and a rain prediction dataset to evaluate the performance of the proposed framework and algorithms. We compare the performance of the proposed framework with three baseline classification methods including K-nearest neighbors, decision trees, support vector machine, and artificial neural networks. The classification accuracy obtained by the proposed framework is comparable with the baseline classification methods.