# Random walk learning

**URL:** <https://discourse.numenta.org/t/random-walk-learning/3967>\
**Category:** Lounge\
**Created:** [May 31, 2018, 2:17am UTC](https://discourse.numenta.org/t/random-walk-learning/3967 "2018-05-31T02:17:45Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sean\_O\_Connor](https://avatars.discourse-cdn.com/v4/letter/s/34f0e0/32.png) [@Sean\_O\_Connor](https://discourse.numenta.org/u/Sean_O_Connor)\
**Post date:** [May 31, 2018, 2:17am UTC](https://discourse.numenta.org/t/random-walk-learning/3967/1 "2018-05-31T02:17:46Z")

</div>

Here is an interesting paper from quite far back about random walk learning:  
[https://arxiv.org/abs/adap-org/9305002](https://arxiv.org/abs/adap-org/9305002)  
It was “discovered” 2 or 3 years ago that neural networks don’t really suffer from a local minimum problem, with enough dimensions local minimums are converted into saddle point problems that weak optimizers (ie. BP) can solve if given time. From the paper they already knew that in 1993!

---

<div class="post-metadata">

**Author:** ![sebjwallace](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/sebjwallace/32/6411_2.png) [@sebjwallace](https://discourse.numenta.org/u/sebjwallace)\
**Post date:** [May 31, 2018, 3:48pm UTC](https://discourse.numenta.org/t/random-walk-learning/3967/2 "2018-05-31T15:48:09Z")

</div>

It reminds me of this study. [http://eplex.cs.ucf.edu/publications/2012/soltoggio-nn12](http://eplex.cs.ucf.edu/publications/2012/soltoggio-nn12)