7 ms·

I recently explained my personal beliefs around how you square free-will and determinism (and subsequently consciousness) to GPT-4 and it told me this was the m

I recently explained my personal beliefs around how you square free-will and determinism (and subsequently consciousness) to GPT-4 and it told me this was the more formal name for it. I posited that if you can observe and reconstruct the entire state of a complex system then you can predict future states- score one for determinism and no free will. But, we know there exists places that we cannot directly observe or perceive, aka quantum uncertainty, represented by σxσp ≥ ℏ/2 [1]. So based completely in theory, I figure the only way we square FW & determinism, is that free will exists somewhere/in a form we cannot directly observe, and it manifests as tiny influences that add up, in the complex system that is a brain. This is the way more speculative part and it's more fun than anything to think about- it doesn't change the way I live my life buuuut- Folded brains dramatically increase the influence a given region in space-time can have, simply due to the increased number of neurons. So our brains double as an antenna for some unseen influence that manifests through quantum uncertainty. So when I explained this to ChatGPT it told me that OORT was very similar to this, but even the mechanism they use for it seems to be a stretch for me. edit: But I do think that in order for neural networks to become anything other than outwardly really really good approximations of human minds, there needs to be a way to introduce a small amount of genuine randomness into their calculations, without utterly breaking them. I could see early attempts at doing this causing a form of LLM schizophrenia because the neural network wasn't resilient enough to the induced error. [1] the standard deviation of position σx and the standard deviation of momentum σp is greater than or equal to half the reduced planck's constant

im3w1l 2y ago
We don't actually know if quantum physics has real randomness or not. Quantum collapse is an unsolved problem. > I could see early attempts at [introducing randomness] causing a form of LLM schizophrenia because the neural network wasn't resilient enough to the induced error. 1. This is actually already done. Temperature parameter controls amount of randomness. 2. Neural networks are quite noise resistant.
- Filligree 2y ago
  The temperature parameter doesn’t introduce any noise into the network evaluation. Typically, what happens is that the network outputs a set of possible tokens with different probabilities, and a sampler picks from the top possibilities. Temperature determines how spiky its pick is; at zero it’ll always pick the top option.
ted_dunning 2y ago
You can get to this conclusion more directly by noting that computational complexity of any Turing simulator of anything more than a trivial system increases very fast as the precision of the initial conditions for the simulation increases. Even the shift map exhibits this phenomenon. This can be an even more severe boundary for prediction than the actual measurement accuracy. In the discussion about determinism vs free will, this leaves us with the situation that we can predict what somebody will do even if we assume perfect measurements, but will only be able to produce this prediction after the fact except for very short term predictions.
- dist-epoch 2y ago
  Stephen Wolfram calls this computational irreducibility.
bbor 2y ago
Why would “my decisions are determined by sub-nuclear divine dice rolls” be any closer to free will than “my decisions are determined by algorithms operating on my sensory inputs and memories”? What’s more “free” about introducing that factor?
maxerickson 2y ago
What does it matter why you can't predict the future state of a brain?
- Bloedcoins 2y ago
  If you can't, we have free will. If we can, we don't have free will.
  - maxerickson 2y ago
    I didn't say "whether", I said "why". At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state (precise information anyway, you may have an opinion about the average state).
    - r2_pilot 2y ago
      >At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state Do we not literally predicate our friendships and relationships on being able to predict the future states of minds? How long do you stay friends with the person who randomly shows up or doesn't, to any event you invite them to? Or whose tastes vary unpredictably from day to day, giving you no framework to contextualize them?
      - maxerickson 2y ago
        It's always very entertaining to nitpick a statement that has a caveat by quoting it without the caveat. (No it isn't)
    - Bloedcoins 2y ago
      If why means because there is a real randomness: we have free will. If its just because of current complexlity, we don't have free will. It also implies that we might life this life over and over forever.
king_magic 2y ago
are we really citing ChatGPT in comments now
- XorNot 2y ago
  There is a damn army of people doing this and I have no idea what they think they're contributing. My personal conspiracy theory is it's ground work to set conditions for disinformation campaigns: the "I used an LLM/I used ChatGPT" people are there to make you look less critically at the other comments by giving a small queue that since they don't include those terms they just be more genuine.
  - ruthmarx 2y ago
    > I have no idea what they think they're contributing. I assume they are just young and see no harm in sharing something they thought was interesting. This fad will die out eventually since it's redundant and provides no real value.
- y-c-o-m-b 2y ago
  Sorry, I must be missing something, what's the problem here? I don't see OP citing ChatGPT, just that they were explaining their own belief system to GPT-4 and it responded by "simplifying" OP's beliefs into "orchestrated objective reduction". This is exactly the type of usage I would expect from an LLM; OP didn't use it to inform their decision, but to further examine the belief from another perspective or broaden their questioning around it.
tsimionescu 2y ago
> But, we know there exists places that we cannot directly observe or perceive, aka quantum uncertainty, represented by σxσp ≥ ℏ/2 [1]. > So based completely in theory, I figure the only way we square FW & determinism, is that free will exists somewhere/in a form we cannot directly observe, and it manifests as tiny influences that add up, in the complex system that is a brain. These two things not only don't follow from each other, the first one actually all but refutes the second. First of all, Heisenberg uncertainty affects all physical systems, but clearly not all physical systems are conscious. Second of all, there is no pattern allowed to exist below Heisenberg uncertainty. That is, if you could determine exactly the momentum of a particle, the particle could literally be anywhere in the universe, with equal probability: there is no bias, it wouldn't be more likely to be here or there. So this is pure randomness, there is no "consciousness signal" you could extract from it. Or, to put it another way, if our consciousness was a result of Heisenberg uncertainty, that would mean it's a purely random phenomenon, and every human at every time would be exactly as likely to type the next word in this comment, start running in a random direction, gouge out one eye, or any other thing they are capable of doing. There is, in a very fundamental sense, no way to get patterns or intention out of Heisenberg uncertainty. Besides, the best way to square "free will" with determinism is Compatibilism. Every human is an automaton whose behavior is fully determined by genetic and epigenetic make-up and by everything they've ever learned and otherwise experienced. In a fundamental sense, my whole life's course was determined the moment I was conceived; but still, in any given situation, what I will do is different from someone else might do, because they have a different history and thus different values and biases. There is no magic that allows some "fundamental me" to "choose" how some electro-chemical processes will fire in my brain, any more than I could "choose" to emit electrons from the tips of my fingers. But that doesn't mean that I (the adult I am today) would do the same things Hitler did if I were somehow catapulted into his shoes today.
n4r9 2y ago
You've outlined what I reckon is the appeal of "quantum consciousness". I personally feel that randomness doesn't help to explain free will any more than determinism. I'm more inclined to believe that free will (in the strictest sense) is an illusion.
- carlmr 2y ago
  The problem with this approach is that even if you say that our thinking is non-deterministic because of true random effects on the quantum level, you still have to explain how deterministic calculations on random values make for free will. You still have no influence on it, even if there is randomness involved.
  - cogman10 2y ago
    You also have to explain why will is changed when the brain is damaged. Really hard to justify free will (IMO) when a person's entire personality can be fundamentally altered by a bash to the head. What does "free will" mean if everything that makes you you can be changed with, say, a lobotomy. It is, at best, an illusion and nothing more.
    - ruthmarx 2y ago
      There is no illusion, and brain damage has no bearing on free will. Free will is simply you making a choice, that's it. If you want to argue about what 'you' means, feel free, but it doesn't really change anything here.
      - n4r9 2y ago
        What is "choice"? Is it simply executing one of a set of possibilities? If you take such a general definition of free will, then a slot machine is manifesting it's free will to deny you a payout.
        ruthmarx 2y ago
Spacecosmonaut 2y ago
Randomness just introduces branch points into the linear flow of deterministic states. Since you do not control the branch points or create them, this does not give you free will.
MattPalmer1086 2y ago
Randomness does not give you free will, any more than determinism does. What do you mean by free will?
- lupusreal 2y ago
  Exactly. If determinism is incompatible with somebody's personal meaning of free will, quantum dice rolls are hardly a solution. What they really need is to either find a religion or just shrug off philosophy and get on with their life, behaving as if they have free will even if they can't rationally justify it.
  - IWeldMelons 2y ago
    Quantum dice roll is _the free will_ in this context. So your free will is what sets the dice; as it is extraphysical, it will look like randomness in the physical world.
    - lupusreal 2y ago
      That's not quantum physics, that's just some sort of new-age religion. A new variation on the "brains are antenna for the soul" idea.
jmcqk6 2y ago
This is possibly one way to solve it, but I think there is a simpler way, following causal chains and the laws of thermodynamics. We clearly have systems that can absorb energy for later use - creating a natural "pause" in the causal chain. Each of these pauses create a possible future that is not yet realized. The longer this energy is held, the larger this possibility space becomes. Free will becomes that ability to hold the pause with intention, and then select from the different possible futures that have been created. Determinism does not interfere with this in any way. The causal chains all follow the basic deterministic laws of physics. There is space for choice created by holding energy instead of immediately dissipating it. No quantum mechanics required at all.
samatman 2y ago
> free will exists somewhere/in a form we cannot directly observe John Conway has a rather neat explanation of this in the Strong Free Will Theorem. https://www.ams.org/notices/200902/rtx090200226p.pdf https://www.ams.org/notices/200902/rtx090200226p.pdf Being neat doesn't necessarily mean it's correct, but it's compatible with what we know about physical reality, and solves some otherwise rather tough and paradoxical facts about experienced reality, so I'm a fan.
ruthmarx 2y ago
> how you square free-will and determinism (and subsequently consciousness) I've never seen this as an issue. Even if something is fated, it's still you making that choice. You ate whatever you ate for lunch yesterday. It's already happened. You still made the choice.

im3w1l 2y ago

We don't actually know if quantum physics has real randomness or not. Quantum collapse is an unsolved problem. > I could see early attempts at [introducing randomness] causing a form of LLM schizophrenia because the neural network wasn't resilient enough to the induced error. 1. This is actually already done. Temperature parameter controls amount of randomness. 2. Neural networks are quite noise resistant.

Filligree 2y ago
The temperature parameter doesn’t introduce any noise into the network evaluation. Typically, what happens is that the network outputs a set of possible tokens with different probabilities, and a sampler picks from the top possibilities. Temperature determines how spiky its pick is; at zero it’ll always pick the top option.

ted_dunning 2y ago

You can get to this conclusion more directly by noting that computational complexity of any Turing simulator of anything more than a trivial system increases very fast as the precision of the initial conditions for the simulation increases. Even the shift map exhibits this phenomenon. This can be an even more severe boundary for prediction than the actual measurement accuracy. In the discussion about determinism vs free will, this leaves us with the situation that we can predict what somebody will do even if we assume perfect measurements, but will only be able to produce this prediction after the fact except for very short term predictions.

dist-epoch 2y ago
Stephen Wolfram calls this computational irreducibility.

bbor 2y ago

Why would “my decisions are determined by sub-nuclear divine dice rolls” be any closer to free will than “my decisions are determined by algorithms operating on my sensory inputs and memories”? What’s more “free” about introducing that factor?

maxerickson 2y ago

What does it matter why you can't predict the future state of a brain?

Bloedcoins 2y ago
If you can't, we have free will. If we can, we don't have free will.
- maxerickson 2y ago
  I didn't say "whether", I said "why". At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state (precise information anyway, you may have an opinion about the average state).
  - r2_pilot 2y ago
    >At the moment, you can't predict the future state of my brain for more than one reason, one of which is that you don't have much information about the current state Do we not literally predicate our friendships and relationships on being able to predict the future states of minds? How long do you stay friends with the person who randomly shows up or doesn't, to any event you invite them to? Or whose tastes vary unpredictably from day to day, giving you no framework to contextualize them?
    - maxerickson 2y ago
      It's always very entertaining to nitpick a statement that has a caveat by quoting it without the caveat. (No it isn't)
  - Bloedcoins 2y ago
    If why means because there is a real randomness: we have free will. If its just because of current complexlity, we don't have free will. It also implies that we might life this life over and over forever.

king_magic 2y ago

are we really citing ChatGPT in comments now

XorNot 2y ago
There is a damn army of people doing this and I have no idea what they think they're contributing. My personal conspiracy theory is it's ground work to set conditions for disinformation campaigns: the "I used an LLM/I used ChatGPT" people are there to make you look less critically at the other comments by giving a small queue that since they don't include those terms they just be more genuine.
- ruthmarx 2y ago
  > I have no idea what they think they're contributing. I assume they are just young and see no harm in sharing something they thought was interesting. This fad will die out eventually since it's redundant and provides no real value.
y-c-o-m-b 2y ago
Sorry, I must be missing something, what's the problem here? I don't see OP citing ChatGPT, just that they were explaining their own belief system to GPT-4 and it responded by "simplifying" OP's beliefs into "orchestrated objective reduction". This is exactly the type of usage I would expect from an LLM; OP didn't use it to inform their decision, but to further examine the belief from another perspective or broaden their questioning around it.

tsimionescu 2y ago

> But, we know there exists places that we cannot directly observe or perceive, aka quantum uncertainty, represented by σxσp ≥ ℏ/2 [1]. > So based completely in theory, I figure the only way we square FW & determinism, is that free will exists somewhere/in a form we cannot directly observe, and it manifests as tiny influences that add up, in the complex system that is a brain. These two things not only don't follow from each other, the first one actually all but refutes the second. First of all, Heisenberg uncertainty affects all physical systems, but clearly not all physical systems are conscious. Second of all, there is no pattern allowed to exist below Heisenberg uncertainty. That is, if you could determine exactly the momentum of a particle, the particle could literally be anywhere in the universe, with equal probability: there is no bias, it wouldn't be more likely to be here or there. So this is pure randomness, there is no "consciousness signal" you could extract from it. Or, to put it another way, if our consciousness was a result of Heisenberg uncertainty, that would mean it's a purely random phenomenon, and every human at every time would be exactly as likely to type the next word in this comment, start running in a random direction, gouge out one eye, or any other thing they are capable of doing. There is, in a very fundamental sense, no way to get patterns or intention out of Heisenberg uncertainty. Besides, the best way to square "free will" with determinism is Compatibilism. Every human is an automaton whose behavior is fully determined by genetic and epigenetic make-up and by everything they've ever learned and otherwise experienced. In a fundamental sense, my whole life's course was determined the moment I was conceived; but still, in any given situation, what I will do is different from someone else might do, because they have a different history and thus different values and biases. There is no magic that allows some "fundamental me" to "choose" how some electro-chemical processes will fire in my brain, any more than I could "choose" to emit electrons from the tips of my fingers. But that doesn't mean that I (the adult I am today) would do the same things Hitler did if I were somehow catapulted into his shoes today.

n4r9 2y ago

You've outlined what I reckon is the appeal of "quantum consciousness". I personally feel that randomness doesn't help to explain free will any more than determinism. I'm more inclined to believe that free will (in the strictest sense) is an illusion.

carlmr 2y ago
The problem with this approach is that even if you say that our thinking is non-deterministic because of true random effects on the quantum level, you still have to explain how deterministic calculations on random values make for free will. You still have no influence on it, even if there is randomness involved.
- cogman10 2y ago
  You also have to explain why will is changed when the brain is damaged. Really hard to justify free will (IMO) when a person's entire personality can be fundamentally altered by a bash to the head. What does "free will" mean if everything that makes you you can be changed with, say, a lobotomy. It is, at best, an illusion and nothing more.
  - ruthmarx 2y ago
    There is no illusion, and brain damage has no bearing on free will. Free will is simply you making a choice, that's it. If you want to argue about what 'you' means, feel free, but it doesn't really change anything here.
    - n4r9 2y ago
      What is "choice"? Is it simply executing one of a set of possibilities? If you take such a general definition of free will, then a slot machine is manifesting it's free will to deny you a payout.
      - ruthmarx 2y ago

Spacecosmonaut 2y ago

Randomness just introduces branch points into the linear flow of deterministic states. Since you do not control the branch points or create them, this does not give you free will.

MattPalmer1086 2y ago

Randomness does not give you free will, any more than determinism does. What do you mean by free will?

lupusreal 2y ago
Exactly. If determinism is incompatible with somebody's personal meaning of free will, quantum dice rolls are hardly a solution. What they really need is to either find a religion or just shrug off philosophy and get on with their life, behaving as if they have free will even if they can't rationally justify it.
- IWeldMelons 2y ago
  Quantum dice roll is _the free will_ in this context. So your free will is what sets the dice; as it is extraphysical, it will look like randomness in the physical world.
  - lupusreal 2y ago
    That's not quantum physics, that's just some sort of new-age religion. A new variation on the "brains are antenna for the soul" idea.

jmcqk6 2y ago

This is possibly one way to solve it, but I think there is a simpler way, following causal chains and the laws of thermodynamics. We clearly have systems that can absorb energy for later use - creating a natural "pause" in the causal chain. Each of these pauses create a possible future that is not yet realized. The longer this energy is held, the larger this possibility space becomes. Free will becomes that ability to hold the pause with intention, and then select from the different possible futures that have been created. Determinism does not interfere with this in any way. The causal chains all follow the basic deterministic laws of physics. There is space for choice created by holding energy instead of immediately dissipating it. No quantum mechanics required at all.

samatman 2y ago

> free will exists somewhere/in a form we cannot directly observe John Conway has a rather neat explanation of this in the Strong Free Will Theorem. https://www.ams.org/notices/200902/rtx090200226p.pdf https://www.ams.org/notices/200902/rtx090200226p.pdf Being neat doesn't necessarily mean it's correct, but it's compatible with what we know about physical reality, and solves some otherwise rather tough and paradoxical facts about experienced reality, so I'm a fan.

ruthmarx 2y ago

> how you square free-will and determinism (and subsequently consciousness) I've never seen this as an issue. Even if something is fated, it's still you making that choice. You ate whatever you ate for lunch yesterday. It's already happened. You still made the choice.