TY - JOUR
T1 - Weighted Graph-Based Signal Temporal Logic Inference Using Neural Networks
AU - Baharisangari, Nasim
AU - Hirota, Kazuma
AU - Yan, Ruixuan
AU - Julius, Agung
AU - Xu, Zhe
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2022
Y1 - 2022
N2 - 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 letter, we propose a method that trains neural networks to learn spatial-temporal properties in the form of weighted graph-based signal temporal logic (w-GSTL) formulas. For learning w-GSTL formulas, we introduce a flexible w-GSTL formula structure in which the user's preference can be applied in the inferred w-GSTL formulas. In the proposed framework, each neuron of the neural networks corresponds to a subformula in a flexible w-GSTL formula structure. We initially train a neural network to learn the w-GSTL operators, and then train a second neural network to learn the parameters in a flexible w-GSTL 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.
AB - 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 letter, we propose a method that trains neural networks to learn spatial-temporal properties in the form of weighted graph-based signal temporal logic (w-GSTL) formulas. For learning w-GSTL formulas, we introduce a flexible w-GSTL formula structure in which the user's preference can be applied in the inferred w-GSTL formulas. In the proposed framework, each neuron of the neural networks corresponds to a subformula in a flexible w-GSTL formula structure. We initially train a neural network to learn the w-GSTL operators, and then train a second neural network to learn the parameters in a flexible w-GSTL 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.
KW - Neural networks
KW - weighted graph-based signal temporal logic
UR - http://www.scopus.com/inward/record.url?scp=85122105255&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85122105255&partnerID=8YFLogxK
U2 - 10.1109/LCSYS.2021.3138059
DO - 10.1109/LCSYS.2021.3138059
M3 - Article
AN - SCOPUS:85122105255
SN - 2475-1456
VL - 6
SP - 2096
EP - 2101
JO - IEEE Control Systems Letters
JF - IEEE Control Systems Letters
ER -