Logic Supervised Learning for Time Series - Continual Learning for Appliance Detection

Classifying parts of time series is an important task when it comes to the usage of Artificial Neural Networks (ANN), e.g. for analyzing the power consumption of households. To make it possible to adapt such ANN for Non Intrusive Load Monitoring (NILM) for the household in which they are deployed is crucial but not easy to manage. The Neurosymbolic Artificial Intelligence (AI) approach in this paper makes it possible to do that by combining ANN modules with Probabilistic Logic which is used as a supervise process to check the outputs of the ANN in case of plausibility. This on the one hand filters implausible results out which is helpful for productive usage, on the other hand these post processed results can be used to retrain the network and adapt it for a specific household in a continual learning process.