"brain scans" of artificial intelligence processes

[frauenfelder]

Originally published at: http://boingboing.net/2017/02/20/brain-scans-of-artificial.html

…

[Boundegar]

This seems precariously close to bullshit. Are the pictures just pretty, or do they actually convey useful information?

[kmoser]

Yeah, there's almost no way the relatively random bits in a neural net form such pretty pictures without heavy human intervention.

[Nelsie]

I was going to say, Could someone explain what I'm looking at in words of one syllable?

It looks like an illustration of someone's acid trip, so I can totally buy it as a representation of what's going on in a neural net, but I need some simple explanation of how it's representing that and what it means.

[Wanderfound]

They're a profit-chasing tech startup, not a research group, and their website is absolutely awash in buzzword bullshit. Not a serious thing.

Pretty, though. Although so are the organic ones...

[frauenfelder]

This topic was automatically closed after 5 days. New replies are no longer allowed.