Hacker News Reader

Comment by chongli | Hacker News Reader

If I had to guess: neural networks have to operate on pixel data whereas real neurons don't. Brains and eyes have evolved in tandem. Perhaps what makes them so efficient is that the eyes handle some of the processing as a consequence of their physical shape and characteristics.

Look at the eyes of bees. Very different from our own (and from the cameras we build) and perhaps very specialized to the limited set of tasks that bees carry out?

View on HN

On eyes in general: the retina is not like a camera pixel array, rather, it extracts all kinds of features and detects motion:

"Eye smarter than scientists believed: neural computations in circuits of the retina." Gollisch, Tim, and Markus Meister. Neuron 65.2 (2010): 150-164.

http://www.sciencedirect.com/science/article/pii/S0896627309...

Yes. The retina is not so much a part of the eye as an extension of the brain into the eye.

They say the eyes are the window into the soul, but really we're just looking at each other's brains!

The neurons in the eyes pre-process a huge amount of information converting data from about 126M rods and cones to 1M ganglion cells - a pretty amazing amount of data compression.

Ref: https://www.amazon.com/Visual-Thinking-Kaufmann-Interactive-...

From my limited understanding, CNNs seem to approximate this functionality in image processing tasks, where the convolution layers are akin to the retina extracting info (lines, shapes, etc) and compression (kernel striding and pooling), while the later fully-connected neural network works for categorization or whatnot.

That probably doesn't answer the question of how a bee's eyes work, though.

???

I guess that's true for mammalian eyes, but what about insect eyes? If the bumble brain only hold 200k neurons, how many do their eyes have on board?

~7000 elements per compound eye according to this page:

https://www.google.co.uk/amp/s/brookfieldfarmhoney.wordpress...

That's individual lenses, right? But it says that each lens has its own photoreceptor(s). So that is a multiple of 7000.

So if that's three photo receptors per facet, already each eye is rivaling the neuron count of the brain.

That's much of a topology thing...our brain may be a CPU, but we have neurons throughout much of our body, processing extremely low level sensory inputs into richer information that brains can process at a higher level.

And different species have very different neural topologies: I've heard it described that Octopoda act more like an eight-member swarm intelligences than an intelligent eight-armed creature. Because their neural density is much more diffuse in their arms, they do so much more processing there, and the central brain acts more like a coordination unit than anything.

Monolithic brain or modular microbrains...it's the Torvalds/Tannenbaum debate all over again!

Reply on news.ycombinator.com