News for hackers. Stuff that matters.
Re:A Neural Network in 11 Lines of Python (2015) by ocharles on Thursday October 19, 2017 @04:34AM (#15506349)
3blue1brown is currently doing a terrific series on how neural networks work, which nicely compliments this blog post. https://www.youtube.com/watch?v=aircAruvnKk&list=PLZHQObOWTQ...
Re:A Neural Network in 11 Lines of Python (2015) by amelius on Wednesday October 18, 2017 @02:20PM (#15501549)
I'm wondering, given a random truth-table with N binary variables and 1 binary output, what is (worst case) the smallest network that can learn it? (In terms of number of parameters).
Re:A Neural Network in 11 Lines of Python (2015) by fooker on Wednesday October 18, 2017 @03:07PM (#15501990)
For a 100% success rate, it would have to be of the size of the minimal BDD. If you can allow for some errors, this becomes an interesting problem.
Re:A Neural Network in 11 Lines of Python (2015) by amelius on Wednesday October 18, 2017 @06:12PM (#15503695)
Yes, but that's probably the theoretical minimum.

What I'm talking about is the size that is required so that the neural net can learn it. This may be different.

Re:A Neural Network in 11 Lines of Python (2015) by HalcyonicStorm on Wednesday October 18, 2017 @03:29PM (#15502167)
Can you define BDD, please?
Re:A Neural Network in 11 Lines of Python (2015) by Someone on Wednesday October 18, 2017 @03:37PM (#15502216)
Re:A Neural Network in 11 Lines of Python (2015) by fnbr on Wednesday October 18, 2017 @05:31PM (#15503351)
This is technically true, but I wonder how close you could get to 100% with minimal size. I would expect that you could get somewhere around >98% with a network that's a few megabytes in size.
Re:A Neural Network in 11 Lines of Python (2015) by e19293001 on Wednesday October 18, 2017 @07:26PM (#15504183)
The author is currently writing this book:

https://www.manning.com/books/grokking-deep-learning

Re:A Neural Network in 11 Lines of Python (2015) by dahart on Wednesday October 18, 2017 @04:23PM (#15502629)
Having gone through this tutorial (which is great!) and several others, I'm curious what is a good second step for the casual neural network learner?

It sounds like there is a growing bag of tricks neural network researchers are discovering to make training practical and stable for large data sets.

One example would be using relu activation - whenever I play with it in a simple tutorial like this one, training seems to explode and fail much more frequently, so I'm guessing either I'm missing another step people use, or there are some extra constraints on initial conditions?

Using a Gaussian for activation in my tutorials has tended to be more stable and converge much faster, but I assume there is a huge downside lurking somewhere to having a non-monotonically increasing function?

What are the tricks of the trade that a weekend warrior should investigate?

Re:A Neural Network in 11 Lines of Python (2015) by mholt on Wednesday October 18, 2017 @05:07PM (#15503109)
I co-authored a blog post with my lab that has practical advice for debugging DNNs - some of these tips might be helpful to you? https://pcc.cs.byu.edu/2017/10/02/practical-advice-for-build...
Re:A Neural Network in 11 Lines of Python (2015) by maurits on Thursday October 19, 2017 @04:11AM (#15506241)
Stanford CS231n: Convolutional Neural Networks for Visual Recognition [1]

The assignments are excellent and will let you implement a deephish network from practically scratch, including backprop, optimizers, tuning etc.

[1]: http://cs231n.stanford.edu/index.html

Re:A Neural Network in 11 Lines of Python (2015) by jackthetab on Wednesday October 18, 2017 @07:37PM (#15504250)
You may want to check out Andrew Ng's Deep Learning Specialization over on Coursera. [1] One of the courses is specifically about hyperparameter tuning and another about structuring your project. There is a lot of practical information scattered across all the courses.

Yes, I'm taking the specialization and having a blast with it. :-)

[1] https://www.coursera.org/learn/neural-networks-deep-learning

Re:A Neural Network in 11 Lines of Python (2015) by T_D_K on Wednesday October 18, 2017 @04:53PM (#15502947)
Based on the limited amount of information, I'm assuming that by "training explodes" you mean that your gradient descent never reaches a local minimum. Try lowering your learning rate? You may be "stepping over" the minimum.
Re:A Neural Network in 11 Lines of Python (2015) by ghosthamlet on Thursday October 19, 2017 @12:41AM (#15505524)
this is my APL and J implemention: https://github.com/ghosthamlet/ann.apl
Re:A Neural Network in 11 Lines of Python (2015) by zakoud on Thursday October 19, 2017 @10:28AM (#15508188)
Re:A Neural Network in 11 Lines of Python (2015) by craigching on Wednesday October 18, 2017 @05:57PM (#15503592)
If this blog post interests you, you may be interested in his book that is forthcoming: https://www.manning.com/books/grokking-deep-learning
Re:A Neural Network in 11 Lines of Python (2015) by yujinyuz on Wednesday October 18, 2017 @09:23PM (#15504723)
Nice discussion :)
Re:A Neural Network in 11 Lines of Python (2015) by carlosgg on Wednesday October 18, 2017 @11:03AM (#15499678)
Re:A Neural Network in 11 Lines of Python (2015) by KennyCason on Wednesday October 18, 2017 @02:09PM (#15501431)
Quite a while ago, not sure if that's a problem. It's a pretty great write up and worthy of posting again. For me, this was the first time seeing it.
Re:A Neural Network in 11 Lines of Python (2015) by SeanDav on Wednesday October 18, 2017 @02:13PM (#15501484)
You are both right. It is worthwhile knowing that something has been posted before, in order to review old comments if interested and good content is always worth repeating.

In any case, reposting is allowed, for several good reasons, which have been discussed in the past.

Re:A Neural Network in 11 Lines of Python (2015) by KennyCason on Wednesday October 18, 2017 @05:30PM (#15503338)
Indeed, I had not considered that. I may have been used to people saying that links shouldn't be posted twice. The extra context is in fact nice!
Re:A Neural Network in 11 Lines of Python (2015) by traverseda on Wednesday October 18, 2017 @02:13PM (#15501477)
It's nice to have the context of past discussions.
Re:A Neural Network in 11 Lines of Python (2015) by KennyCason on Wednesday October 18, 2017 @05:29PM (#15503328)
Ah, very good point. I hadn't considered that. :)
Re:A Neural Network in 11 Lines of Python (2015) by dang on Wednesday October 18, 2017 @03:24PM (#15502120)
After a year or so it's ok: https://news.ycombinator.com/newsfaq.html. Or if the story hasn't had major attention yet.
Re:A Neural Network in 11 Lines of Python (2015) by SloopJon on Wednesday October 18, 2017 @04:15PM (#15502544)
Perhaps a 2015 label would be appropriate?
Re:A Neural Network in 11 Lines of Python (2015) by dang on Wednesday October 18, 2017 @04:17PM (#15502552)
Doh yes added thanks!