This message was deleted.

we can definitely attest to a difference in ease of tuning, but the lift against xgboost is significant (at least in our domain) and persists across ~95% of all of our ~1k forecasting tasks

XGBoost is all about creating features. Could this be the issue?

we ran both with and without tsfresh features, same finding held up.

Nice there is also new one for features could be interesting for team https://github.com/predict-idlab/tsflex

awesome, will give this a try 👍

Did DL win on global models (multi series) or univariate as well?

yah for clarity this is all for multioutput single series using exogenous variables