# How the backward() works for torch Variable?

**URL:** <https://discuss.pytorch.org/t/how-the-backward-works-for-torch-variable/907>\
**Category:** Uncategorized\
**Created:** [March 6, 2017, 10:47am UTC](https://discuss.pytorch.org/t/how-the-backward-works-for-torch-variable/907 "2017-03-06T10:47:04Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![jekbradbury](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/jekbradbury/32/14_2.png) [@jekbradbury](https://discuss.pytorch.org/u/jekbradbury)\
**Post date:** [March 6, 2017, 6:28pm UTC](https://discuss.pytorch.org/t/how-the-backward-works-for-torch-variable/907/3 "2017-03-06T18:28:10Z")

</div>

Note that if you’re calling it on a loss/cost variable, you don’t need to provide an argument since autograd will assume a tensor of ones.

---

_[View the full topic](https://discuss.pytorch.org/t/how-the-backward-works-for-torch-variable/907)._