Hacker News new | ask | show | jobs
by carsongross 3738 days ago
Conjecture (half serious): we won't get strong AI until we figure out why we need to sleep.
9 comments

Begins to sound like a science fiction novel where a bunch of sleep researchers discover that far from being 'down time' sleep is actually a period where our neural hardware is being used by beings from another dimension to perform some nefarious calculation. Or if Douglas Adams were still alive a P2P version of Deep Thought.
Relevant short story by Issac Asimov: https://en.wikipedia.org/wiki/Hostess_(short_story)
My (semi-serious) theory: Sleep is us logging off of this MMO game we play.

Our "real" bodies (minds) need some downtime from the game.

So this explains why real AI will never happen: This reality is not where our consciousness lies.

Wouldn't they just pause the game though? In which case we'd never notice.
Each person logs off at a different time, it's one game for everyone not a single player game.
Probably related. Mammals dream, and we humans experience time in dreams much faster than the actual passage of time. Dreaming is likely the same mechanism as 'experience replay' used in AI reinforcement learning. We're just training our neutral networks using minibatches of our experiences from our waking hours.
"Likely" is not the word I would go for. "according to my wild guess" is a more accurate term.
I've heard it before and it's not their wild guess but a seriously discussed idea. It's just that it's hard to set up experiments to test these sort of ideas.
It can be a seriously discussed wild guess. It's not like there's some dichotomy to be satisfied there.
I read somewhere a conjecture that sleep for human brains is analogous to regularization schemes (such as dropout) for neural networks. Dropout helps the net avoid overfitting; nets that are overfitted have a difficult time distinguishing noise from signal. Sleep for mammals might serve a similar purpose, to help the brain avoid overfitting to noise (imagine if you had difficulty distinguishing dreams from reality).
Yea, I would suspect this is true as well. Upenn had some sleep studies that were looking this idea, I'n not sure how mature the research was/is though.
Couldn't it go the other way, too? With weak AIs, we get a sandbox of rough sketches of consciousnesses with which to experiment. Some optimisation here or there which looks uncannily like sleep may shine light on the biological version.
I think you guys are interepreting carsongross a little too literally. The bigger point is we understand very little about sleep - we understand even less about how neurons learn and how the brain functions.
Might just be recalculating weights of the Neural Net for learning which can't be done with the system "on line" or concious.

Would make sense with the learning connection and memory.

My hypothesis is that all RNNs (and in general complex dynamical systems) need to be reset periodically. If run for too long without resetting, they tend to get stuck in strange states, blow up, or cease activating. You can see this effect by running a generative RNN model for a long time - eventually the output is garbage.

Under this model the next obvious question is why it takes so long to reset the brain's state. Maybe it can be done faster.

Maybe, maybe not. If it turns out it's only a biological cause (waste disposal, tired neurons, ..??), this would not be the case. But good point.
But not before we determine whether salt is bad for us.
I just find it the most amazing thing. 8 hours a day, we all do it, and we don't know why. It's akin to having part of the globe unexplored.

Anyone got any recommendations on lay-person books on sleep/subconscious?