# What's the meaning of the bias?

**URL:** <https://discuss.pytorch.org/t/whats-the-meaning-of-the-bias/13095>\
**Category:** vision\
**Created:** [January 31, 2018, 11:35am UTC](https://discuss.pytorch.org/t/whats-the-meaning-of-the-bias/13095 "2018-01-31T11:35:13Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![11155](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/11155/32/3330_2.png) [@11155](https://discuss.pytorch.org/u/11155)\
**Post date:** [January 31, 2018, 11:35am UTC](https://discuss.pytorch.org/t/whats-the-meaning-of-the-bias/13095/1 "2018-01-31T11:35:13Z")

</div>

In convolution, nn.Conv2d(in\_channels, out\_channels, kernel\_size=3, stride=stride, padding=1, bias=False)  
What’s the meaning of the bias?

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [January 31, 2018, 11:56am UTC](https://discuss.pytorch.org/t/whats-the-meaning-of-the-bias/13095/2 "2018-01-31T11:56:58Z")

</div>

The bias is an additive parameter in the convolution.  
It’s like the `b` in `f(x) = w*x + b`. If you set `bias=False`, you will drop the `b` term, which might make sense in some cases, e.g. if the next layer is an affine BatchNorm layer.  
Each kernel has an own bias term.

However, I think the concept is way better described in [Stanford’s CS231n](http://cs231n.github.io/convolutional-networks/).