How can Anomaly Detector recover from huge spikes and dips?

Hi all. I have switched from Elastic Stack to OpenSearch just to be able to use Anomaly Detection for free. It seems to work. But I have a big question.

Is there any way to tell an Anomaly Detector to disregard a huge spike or dip after the fact? I’ve already seen cases where something we’re tracking goes up or down orders of magnitude due to some technical failure somewhere. And it seems to report it. But I DON’T want it to regard that as a new normal, or factor it in to expected future behavior.

I know it will return to normal after a while, but that could take weeks, right?

Any thoughts? Thank you! :slight_smile: