Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why Is the Human Brain So Efficient? (2018)

nautil.us

Post reply on HN
Print view View original

71–80 of 173 posts

Re: Why Is the Human Brain So Efficient? (2018)

#71

Post by jcims »

stared wrote: post #39

I wouldn't say that human brain is that efficient (per volume). Compare and contrast with the brain of rats or Corvidae: https://www.youtube.com/watch?v=ZerUbHmuY04 .

It's not even a good example. Humans are about the least physically agile vertebrates on the planet.

Think of a fruit fly. It can walk, fly, forage for food, mate, etc. The entire critter has a mass of .2mg and their brains have ~135k neurons. Making the horrible assumption of linear power scaling, that's one microwatt.

Re: Why Is the Human Brain So Efficient? (2018)

#72

Post by GuB-42 »

papito wrote: post #33
sddfd wrote: post #14

I feel uncomfortable at the ubiquitous, silent assumption that what is marketed as AI is a computer implementation of a brain. I see how the term neuronal network reinforces this believe, but we (especially the researchers among us) should allow for the possibility that we are missing something.

Neural networks also have no ability to create new information based on their own mistakes. What is a mistake? When does something look "off" but still very interesting? For example, you can feed a neural net all the recipes of burgers to create a perfect burger. Great. But how does the same net invent the burger? The burger, like many foods or accidental art, was invented as a result of scarcity, circumstance, exper…

GAN can do that. For example, AlphaZero invented strategies for the game of go from nothing but a random number generator and the rules of the game. As for perfection neither go nor chess AIs play perfectly, and they can still beat the best human players.

Of course, an AI intended to play go isn't going to invent the burger. But I see no reason why, given a list of ingredients, their properties and a model of what human enjoy eating, a neural network couldn't invent the burger.

Creating a new recipe is just an optimization problem at its core.

Re: Why Is the Human Brain So Efficient? (2018)

#73

Post by marcosdumay »

cmehdy wrote: post #66
The_rationalist wrote: post #23

Earlier quoted context omitted.

Comparing Human Brain with a CPU is misconception. no it is not. Yeah architecturally they are very different and CPU are arguably more programmable / general and less efficient. What does matter is whether CPUs are theoretically able to achieve all the things that a brain can do (and even more) And indeed CPUs as turing complete, programmable machine are a strict superset of what brains can do. The gap between what…

> CPUs as turing complete, programmable machine are a strict superset of what brains can do In what way can this be proven? It's very tempting in an era of tech-centered growth to think of computers as the solution to everything, but we are barely even beginning to understand the brain. We know computers fairly well and can talk about them, but how can we make such a claim when we don't know the other thing we're tal…

> In what way can this be proven?

Proven? Nothing in science is ever proven.

But on half a millenium we have failed to find anything that can't be simulated by math, and Turing completeness means a computer can simulate anything that can be simulated by math. We also can simulate all the smallest components of a brain.

At this point the claim that math can not simulate it is highly extraordinary.

Re: Why Is the Human Brain So Efficient? (2018)

#74

Post by jacobr1 »

cmehdy wrote: post #66
The_rationalist wrote: post #23

Earlier quoted context omitted.

Comparing Human Brain with a CPU is misconception. no it is not. Yeah architecturally they are very different and CPU are arguably more programmable / general and less efficient. What does matter is whether CPUs are theoretically able to achieve all the things that a brain can do (and even more) And indeed CPUs as turing complete, programmable machine are a strict superset of what brains can do. The gap between what…

> CPUs as turing complete, programmable machine are a strict superset of what brains can do In what way can this be proven? It's very tempting in an era of tech-centered growth to think of computers as the solution to everything, but we are barely even beginning to understand the brain. We know computers fairly well and can talk about them, but how can we make such a claim when we don't know the other thing we're tal…

The argument isn't "something like, or a little better, than current CPUs can perform everything a brain can," but something more like "a turing machine can perform everything a brain can or more." This is more an ontological exercise, not an empirical one. If you reduce everything to a "black box" model with inputs and outputs, then sure, the mathematical abstractions of theoretical brains and theoretical CPUs have a congruence. Most objections to this seem to resolve around qualia being something not modelable in machines, but I'm skeptical of that claim.

Can an "arbitrarily advanced computer do everything a brain can do?" Empirically, right now, current machines can't but we are talking about "future machines, via line-of-sight extrapolation". Not fundamental leaps in tech, but incremental ones. It seems plausible, but it seems we expand the depths of the complexity of the requirements nearly as fast as we advance current capabilities. I don't know, but I'd put my money on the technology catch up.

Re: Why Is the Human Brain So Efficient? (2018)

#75

Post by sa1 »

tegeek wrote: post #15
est31 wrote: post #9

One kind of efficiency which hasn't been talked about is the energy loss of things like state switching and keeping the current state enabled. I think that brains build on much more efficient primitives than the silicon transistors computer chips use and thus can perform far more computations for far less energy than a desktop CPU. Another difference between CPUs and brains is that brains are much less general purpos…

Comparing Human Brain with a CPU is misconception. In the past when we didn't have digital computers we used to compare Brain with other machines. And now with a CPU. A Brain from a primitive neuron to higher level is not comparable to any machine at all including the CPU.

Computers are mathematical concepts, Turing machines being one such concept. Whether computers are implemented using silicon, or oil, or using neurons, it doesn't really matter as we have a mathematical framework for describing abstract machines, and we can determine what is a machine, and what is not.

We did not have this mathematical framework before the age of Turing, Church, Russel, et al.

This doesn't mean that brains are very similar to CPUs, they are not, just like they were not similar to mechanical machines before.

Yet we do now have a way of studying the similarities they have.

Re: Why Is the Human Brain So Efficient? (2018)

#76

Post by bluedino »

est31 wrote: post #9

One kind of efficiency which hasn't been talked about is the energy loss of things like state switching and keeping the current state enabled. I think that brains build on much more efficient primitives than the silicon transistors computer chips use and thus can perform far more computations for far less energy than a desktop CPU. Another difference between CPUs and brains is that brains are much less general purpos…

It's like comparing a human to a horse. A horse can run very fast or pull a wagon. But a horse can't work on plumbing, or knit.

Re: Why Is the Human Brain So Efficient? (2018)

#77

Post by coreyp_1 »

supergarfield wrote: post #67
coreyp_1 wrote: post #63

Earlier quoted context omitted.

Regarding the Church-Turing thesis, it is exactly that, just a thesis. Again, akin to P vs NP. It seems to hold for most cases, but is not proven. The reason that it's difficult to apply in regards to the brain is that we don't exactly know how the brain is computing... or if it "computes" at all! To my knowledge, we don't have a model of computation for consciousness, emotion, free will, Etc. Perhaps these are bette…

As far as I understand, the prevailing opinion is that the brain is a physical object and that its operation does not involve currently-unknown laws of physics (because we have a good understanding of what happens at the scale of an entire atom or above). A Turing machine can run a simulation based on such physical laws to any desired level of precision (which is enough, because as mentioned in TFA, processes in the…

"any desired level of precision" is actually the issue. The moment you choose a level of precision, you cease being accurate (at that level). If you make the argument that a TM has infinite memory, and can therefore represent an infinite precision, then I would counter that our current defintion of a TM requires a finite tape alphabet (and finite number of states), which is part of the TM's known computational limitations. And, of course, the moment that you use any finite set of symbols to represent an infinitely precise value, you fall into the problem that the set of real numbers has a larger cardinality than the set of possible turing machines (again, simple proof via diagonalization).

It is possible that the brain's imprecision (I would argue that "inconsistency" might be a better word) is a requirement of it's computational ability. Again, we haven't defined how the brain computes, nor do we have a model for explaining its computation, encoding or representation of knowledge, or emergent behavior. We have observed phenomena related to some of these things, but we are far from understanding it. It may be that the computational processes are dependent on the surrounding environment. We know that the biological processes are influenceable by the physical world, but we do not know much about how these external forces affect, limit, or are required for, the process of brain computation.

The quantum world may play a part in consciousness (or no, we don't know). Non-determinism may play a part. It is possible that, in order to simulate a brain, one would have to simulate the entire universe around it in order to predict the behavior... meaning that it may well require a universe to perform the simulation.

Which brings us to a related theory of whether or not we are living in a simulation, but I digress... :)

Re: Why Is the Human Brain So Efficient? (2018)

#78

Post by EmilioMartinez »

kristopolous wrote: post #69

They aren't the same thing. They are different classes of objects, different tasks. This comparison is kind of silly. I'd hate my computer to have the memory accuracy or the computational accuracy of my brain. I'd hate to have the creativity and inspiration of a computer. Delete being such a nontrivial operation is probably a good thing for humans. Copy being imperfect probably has something to do with the phenomenon…

I don't see how any of that makes the comparison "silly". It's not like we have so many instances of computer paradigms to go around comparing.

Re: Why Is the Human Brain So Efficient? (2018)

#79

Post by sharpneli »

SomeoneFromCA wrote: post #51
sooheon wrote: post #43

Earlier quoted context omitted.

Biology is the ultimate legacy software running on one of the oldest platforms ever developed, the organic compounds. It is literally a giant genetic algorithm to write instructions (DNA) for manufacturing molecular machines (proteins) that interact with each other in an extremely complex graph of relations (protein pathways, i.e. control flow).

This a very simplistic view, based on assumption that the world is discrete. The whole idea of software relies on the concept of digital computer, a discrete machine. The world might indeed be analogous and real numbers might actually exist.

If world did run on real numbers that we could harness for computation I would be more than happy, because using those we would be able to perform hypercomputation. See https://en.m.wikipedia.org/wiki/Real_computation

However this is forbidden by Bekensteins bound, so unless modern physics is horribly broken it’s ruled out at least in any sense visible to us even in principle.

Re: Why Is the Human Brain So Efficient? (2018)

#80

Post by derefr »

kristopolous wrote: post #69

They aren't the same thing. They are different classes of objects, different tasks. This comparison is kind of silly. I'd hate my computer to have the memory accuracy or the computational accuracy of my brain. I'd hate to have the creativity and inspiration of a computer. Delete being such a nontrivial operation is probably a good thing for humans. Copy being imperfect probably has something to do with the phenomenon…

> memory accuracy

There are individuals with very good memories for all sorts of things, who seem to manage to reconsolidate their memories near-losslessly (at least within the confines of the mental schema they organize said memories into.) Surgeons with anatomy, lawyers and judges with case-law, etc.

At this point I’m convinced that the lossy method humans intuitively reconsolidate memories with, isn’t so much a feature of our mental architecture, as it is a part of the “operating system” we build up on top of our mental architecture—i.e. it’s a skill, something we can learn (or accidentally invent) a better approach to.

> computational accuracy

We compute ratios with extremely high accuracy/precision. Just look at a professional billiards player.

We don’t have a good mind for integer math; but you can translate most integer math problems into ratio problems, and then they become intuitively solvable to humans. (This is basically what geometry is.)

Post reply on HN