Is there any good papers / surveys on transfer learning / domain adaptation for time series forecasting?

I am mostly working with LSTM recurrent neural networks, but any other ML-based approach should be an interesting read.