# Weight Agnostic Neural Networks
**URL:** https://discourse.numenta.org/t/weight-agnostic-neural-networks/6467
**Category:** Current Research
**Tags:** live, journal-club
**Created:** [August 15, 2019, 4:13pm UTC](https://discourse.numenta.org/t/weight-agnostic-neural-networks/6467 "2019-08-15T16:13:13Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@rhyolight](https://discourse.numenta.org/u/rhyolight)
#### Post date: [August 15, 2019, 4:13pm UTC](https://discourse.numenta.org/t/weight-agnostic-neural-networks/6467/1 "2019-08-15T16:13:13Z")
A Numenta Journal Club and live paper review:
Weight Agnostic Neural Networks:
> **[Weight Agnostic Neural Networks](https://arxiv.org/abs/1906.04358)**
>
> Not all neural network architectures are created equal, some perform much
> better than others for certain tasks. But how important are the weight
> parameters of a neural network compared to its architecture? In this work, we
> question to what extent...
Interactive (more updated) version:
> **[Weight Agnostic Neural Networks](https://weightagnostic.github.io/)**
>
> Networks that can already (sort of) perform tasks with random weights.
The is a fairly new paper (Jun 11) that reinforces the implication of the Lottery Ticket Hypothesis: that the weights of a network don’t matter as much as people think and there’s a lot of importance in the structure of a network.
---
### Author:  [@rhyolight](https://discourse.numenta.org/u/rhyolight)
#### Post date: [August 15, 2019, 4:22pm UTC](https://discourse.numenta.org/t/weight-agnostic-neural-networks/6467/2 "2019-08-15T16:22:07Z")
---
### Author:  [@rhyolight](https://discourse.numenta.org/u/rhyolight)
#### Post date: [August 16, 2019, 5:20pm UTC](https://discourse.numenta.org/t/weight-agnostic-neural-networks/6467/3 "2019-08-16T17:20:52Z")
We are live.