> 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/sep/learning-to-stop-while-learning-to-predict.md).

# Learning to stop while learning to predict

## Motivation

* Task-imbalanced meta learning: different tasks need different numbers of gradient steps for adaptation 
* Deep learning based algorithms usually have a fixed number of iterations in the architecture.

could we learn to stop automatically?

## Overview

![](/files/-MIHn7TXAJZfUu2vMRR1)

* predictive model $$\mathcal{F}\_\theta$$ : transforms the input x to generate a path of states $$x\_1, x\_2, ...,x\_T$$ 
* stop policy $$\pi\_\phi$$ : sequentially observes the states $$x\_t$$ and determines the probability of stop at layer $$t$$ 
* variational stop time distribution $$q\_\phi$$ : stop time distribution induced by stopping policy $$\pi\_\phi$$ 

![](/files/-MII4beOqxNSI_036r-M)

![](/files/-MII4etkJerQx9ot0qbV)

## Reference

* <https://icml.cc/Conferences/2020/ScheduleMultitrack?event=6279>
* <https://icml.cc/media/Slides/icml/2020/virtual(no-parent)-14-17-00UTC-6279-learning_to_sto.pdf>
* <https://www.youtube.com/watch?v=3VqzhP44Eic>