> For the complete documentation index, see [llms.txt](https://lichangbin.gitbook.io/paper_notes/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://lichangbin.gitbook.io/paper_notes/oct/reviving-and-improving-recurrent-back-propagation.md). # Reviving and Improving Recurrent Back-Propagation ## Motivation the original BPTT is not efficient, because of the inverse matrix instead of using this, this paper use Neurman series to approximate the inverse matrix to reduce the cost. similar to the hyperparameter optimization paper ## Reference *