Intelligent Time-Series Processing

We propose a formal interpretation framework that allows a human user to interrogate a cyber-physical system using temporal logic queries. The method formulates interpretation as a temporal logic inference problem and introduces a Gaussian-process-based active learning algorithm to obtain probably approximately correct solutions efficiently.

We formalize semantic fault diagnosis for IIoT systems as the task of learning signal temporal logic formulas directly from time-series data. To address combinatorial explosion, we propose an agenda-based, reinforcement-learning-enabled search strategy that discovers scalable and interpretable fault specifications from industrial sensor data.

We present the problem of semantic inference for cyber-physical systems, aiming to translate system behaviors into signal temporal logic specifications automatically. To address combinatorial explosion in formula search, we propose an agenda-based computation framework and formulate semantic inference as a Markov decision process solved with reinforcement learning.