This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Point of order: that a model can, under certain conditions, replicate aspects of behaviour does not in any way indicate that the brain operates using those principles. Every time we make a new mechanical process, people decide that's how the brain works on some fundamental level. Switchboards, electrical fields, computer logic. Last decade it was Reinforcement Learning (S.A. still asserts this is settled science in his posts and in academic neuroscience it's popular but far from settled), this decade it's autoencoders (predictive coding) or LLMs.
I'm not saying you're wrong, I'm just saying we're far, far off proving it's true.
FWIW I agree on the core point that statistical analysis is batting much stronger than formal logic, but against that you do have to weight the fact that for some reason LLMs need 10,000x the amount of data to go the same distance. Probably some combination of specialised circuits and us missing serious statistical factors IMO.
Yeah, and this is mostly what people mean when they say this:
Victorians or Edwardians who imagined the mind as comprised of millions of tiny gears and levers like Babbage’s computer had obviously dissected brains, they mostly recognized it wasn’t literally like that, it was always more about the kind of operation than the exact physical mechanism.
Is it settled science that the human mind is a specific kind of ML model that maps to a category computer scientists would recognize? Obviously not, it’s unlikely something that evolved by circumstance on a completely different kind of computer we don’t understand would be comparable in that way. But it probably is a neural network or something comparable to a neural network of some kind.
Probably, but that's a biiiiig space. Neurons are weird. Like, really weird. As with junk DNA there's a huge chunk of them which exist purely to mediate the interaction of other ones, they change their behaviour based on the +ve/-ve sign of the timing between electricial signals, all the individual connections are running surprisingly complicated and hard-to-tease-out statistical models of their own. Plus you have all the area specialisation: speech areas, vision, motor, limbic etc. and all the heritage structure from evolution.
It may turn out that these are really a very bad, overtuned method of running a simple system, which is what the RL and the predictive-coding people are hoping for. It might also turn out that the brain is a lot cooler and weirder than we know. I don't think we're even in a position to speculate at the moment.
It might turn out that there are just many ways to do the same thing. It might also be true that all language models, being essentially ‘distillation’ (via training sets which are the sum of human art, writing, knowledge, etc) of our wisdom, can replicate and build on our achievements but also required, in that fundamental sense, the ramshackle evolutionarily derived networks we have in our own heads that were built for the multimodal real world environment biological life inhabits. We’ll know once we have recursive self-improvement and can attempt initial training with extremely primitive datasets.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link