[keghn]

Most of the neural network programs and libraries are Python.   

[keghn]

WebAssembly: 
https://en.wikipedia.org/wiki/WebAssembly   

Firefox 58 Bringing Faster WebAssembly Compilation With Two-Tiered Compiler: 
https://www.phoronix.com/scan.php?page=news_item&px=Firefox-58-2-Tiered-Compiler   

[Zero]

I understand danlovy, sticking to what you know lets you focus on the content of your project, and that's a good thing.

On the other hand, if you haven't done any coding for many years, you're probably far away from what C++ has become. It has evolved, and modern C++ has its own learning curve.

Anyway, C++ can be ported to JS easily thanks to tools like emscripten, so if one day you change your mind, you know you can have the best of both worlds!

[danlovy]

Thanks.  My goal was actually not to code but to test some ideas.  Here is a LinkedIn article I wrote describing this effort.

https://www.linkedin.com/pulse/industrial-strength-evolution-genetics-ai-dan-lovy/

I'm recruiting a team to take this to the next level.

Dan Lovy

[ivan.moony]

Hi Dan Lovy

I'm wondering what would this next level specifically be?

[danlovy]

Well, a lot of NN technologies are finding their way into commercial apps.  All the new fangled face recognizers use them.   The will be in a self driving car in your neighborhood soon.

While they take their cue from nature, they are really an exercise in some math wizardry.   Try this
http://20q.net/ 

It does a pretty good job of playing 20 questions, however, it really doesn't know anything.

It's a little too involved to go into deep detail on what I'm thinking about however, my goal is to take genetic algorithms and provide a platform that better simulates how more powerful systems might be created.

dan

[danlovy]

3rd Episode, a better maze solver