A suggestion to add a pre-training method article

[realwangjiahao]

Thank you for summarizing these articles. It is very helpful to my study of time series.

I am very interested in the Pretrain & Representation part, and your summary is also very comprehensive.
I recently read an article about pre-training. The author proposes TimeMAE, a novel self-supervised paradigm for learning transferrable time series representations based on transformer networks. 
Could you please read this article and add it to the Pretrain & Representation part？
You can find this article at [https://arxiv.org/abs/2303.00320]

With best regards.

[ddz16]

Thank you for your reminder, I have added this paper.