Systems and methods for detecting and accommodating state changes in modelling

One or more non-transitory computer-readable media storing instructions, which when executed by one or more hardware processors, cause at least one computing device to perform operations comprising:

• receiving a time-series signal that includes a sequence of values captured by one or more computing devices over time;
• generating, within at least one of volatile or non-volatile storage, a representation of the time-series signal;
• identifying a set of one or more state changes within the representation of the time-series signal;
• selecting, based at least in part on at least one state change in the set of one or more state changes, a subset of the values from the sequence of values to train a model as if there were no state changes in the time-series signal, wherein the subset of values excludes one or more values that occurred prior to the state change;
• training, by at least one computing device, a model using the selected subset of values from the sequence of values;
• and
• generating, within at least one of volatile or non-volatile storage of at least one computing device, an analytical output using the trained model.