Hey there, just joined, not sure if we’re supposed to present ourselves in this slack but I’ll do it anyway 🙂 I work on an anomaly detection system called ThirdEye. We use a bunch of forecasting-based method for anomaly detection, even though I mostly work on matrix profile these days. Just discovered your libs with Statistical vs Deep Learning forecasting methods piece of work, and will definitely dig it. Thanks for the awesome work!

Hi @Cyril de Catheu! Thanks for joining. @Cristian (Nixtla) did some work on anomaly detection at Amazon; he might be an interesting talking partner. Congrats on Third Eye, looks very exciting. I would like to say the same about the matrix profile, but I think that thing is above my intellectual capabilities. We are happy to help in way possible.

@Cyril de Catheu Matrix Profile is great but have you checked out Conformal Prediction for anomaly detection?

thanks @Max (Nixtla). I find what you’re doing at Nixtla more complex that matrix profile 🙃 @Valeriy nope, do you have a good reference to start with?

@Cyril de Catheu I do, here https://valeman.medium.com/how-to-detect-anomalies-state-of-the-art-methods-using-conformal-prediction-e02691659412

nice! I spend a lot of time in my company explaining that forecasting-based methods are not good for anomaly detection will dig into KNN-CAD

@Cyril de Catheu spot on forecasting is not great for anomaly detection, Conformal Prediciton and MP are much better. Your might also check more recent papers by Vovk on alrw.net http://alrw.net/articles/32.pdf http://alrw.net/articles/31.pdf http://alrw.net/articles/29.pdf this paper from Fedorova and Vovk powers microsoft azure anomaly detection http://alrw.net/articles/04.pdf