Filter Evolution Using Cartesian Genetic Programming for Time Series Anomaly Detection

Industrial monitoring relies on reliable and resilient systems to cope with unforeseen and changing environmental factors. The identification of critical conditions calls for efficient feature selection and algorithm configuration for accurate classification. Since the design process depends on experts who define parameters and develop classification models, it remains a time-consuming and error-prone task. In this paper, we suggest an evolutionary learning approach to create filter pipelines for machine health and condition monitoring. We apply a method called Cartesian Genetic Programming (CGP) to explore the search space and tune parameters for time series segmentation problems. CGP is a nature-inspired algorithm where nodes are aligned in a twodimensional grid. Since programs generated by CGP are compact and short, we deem this concept efficient for filter evolution and parameter tuning to create performant classifiers. For better use of resources, we introduce a dependency graph to allow only valid combinations of filter operators during training. Furthermore, this novel concept is critically discussed from a efficiency and quality point of view as well as its effect on classifier accuracy on scarce data. Experimental results show promising results which-in combination with the novel concept-competes with state-of-the-art classifiers for problems of low and medium complexity. Finally, this paper poses research questions for future investigations and experimentation.