Re: [agi] Do Neural Networks Need To Think Like Humans?

I agree too. That the edges are key to AGI. But it just shows how flexible NN can be.   My AGI model and theories parallels in a lot of ways to Korrelan's. Hello Korrelan. But there are great difference in other parts of my model and Korrealan's. So yes in my work i am using a attention

Re: [agi] Do Neural Networks Need To Think Like Humans?

@Matt - I agree. This is a AGI ocular input module, at the end of the video you can see the results from using a polar retina, high resolution fovea and the peripheral degrading lower resolutions.  There are videos showing the negation of rotational/ scale invariance using the same technique.

Re: [agi] Do Neural Networks Need To Think Like Humans?

What this 2 minute video seems to show is that neural networks need to see more like humans to detect shapes rather than just textures. Our visual cortex detects lines, edges, and other regions of high contrast in several layers. But we see shapes by moving the fovea along a path connecting these

Re: [agi] Do Neural Networks Need To Think Like Humans?

'How neural networks learn' - Part III: The learning dynamics behind generalization and overfitting: https://www.youtube.com/watch?v=pFWiauHOFpY -- Artificial General Intelligence List: AGI Permalink:

Re: [agi] Do Neural Networks Need To Think Like Humans?

This was really interesting. Jim Bromer On Sat, Mar 9, 2019 at 10:09 AM wrote: > Do Neural Networks Need To Think Like Humans?: > > > https://www.youtube.com/watch?v=YFL-MI5xzgg > *Artificial General Intelligence List * > / AGI / see discussions

Re: [agi] AI is too important...

> Computer science (CS) is so wrong about AI I probably agree, but I'd like to hear why exactly you think that? The term "AI" is an open term after all, that will be redefined as new ways of AI are found. On Sun, 10 Mar 2019 at 01:38, A.T. Murray wrote: > Computer science (CS) is so wrong