verygoodname

> If you know what a Markov chain is then you must also know that modern language models are nothing like Markov chains. Just as an example, a Markov chain can't do causal reasoning or correctly solve unseen programming puzzles, the way GPT-3 can.

First, GPT-3 can hardly do "causal reasoning" beyond exploiting regularities in the content it has been trained on. That's why you get "interesting causal reasonings" such as "the word 'muslim' co-occurs a lot with 'terrorist', so these things must be causally related".

Second, just because a more complicated version of a Markov chain (i.e. a probabilistic language model) can do things that a first-order Markov chain cannot does not mean that it is qualitatively different: both things are nothing more than simple mathematical models (linear algebra + sprinkles). A polynomial model can do things than a linear model cannot, but it is no closer to consciousness and self-awareness, as far as I can tell.

My point is... a mathematical model is just that (a model) and, as such, cannot be "self-aware".

Humans, as cybernetic agents embedded in some environment (from which they get sensory input and with which they can interact), have agency and, as such, can display the property of "self-awareness". Models, by themselves, cannot.

Humans may contain systems within them that resemble a GPT-3 language model (or a Markov chain model), and such "language models" may even be required for self-awareness (it's not obvious, but let's assume it's true).

*Still*, it is not the language model itself that is self-aware, but the agent that is using the model.

A GPT-3 model, by itself, cannot be self-aware, because it not much more than a applying some linear algebra operations on numbers (just like e.g. a Markov chain language model). An agent containing (among many other things) something that could be approximated by a GPT-3 model within, may.

> As for self-awareness, your brain is an N x N -matrix in the same sense as an ANN, so surely it must be possible for one to be self-aware? Not claiming that GPT-3 is, of course.

No, it is not. My brain is an analog computation device and not a bunch of numbers. Perhaps you can approximate some aspects of how it works using numbers, probably using digital computation devices, but that is not what anyone's brain is (neither at the "hardware" level, nor at the "software" level). Also, notice that a brain always exists within a biological agent, and it is the biological agent that is (or may be) self-aware, and not "the brain".

Would you say that a "Markov chain"-type (e.g. Dissociated Press) language model is "self-aware" in any way?

If yes, then... how, exactly? It is basically an N x N matrix of values.

Current language models (GPT et al.), are qualitatively nothing fancier than this: probabilistic models which encode regularity in language and from which you can sample to get "plausible content".

If a bunch of values in a matrix is "self-aware", then I guess GPT can be seen as "self-aware"; if not, then it can't.

My problem is trying to imagine an N x N matrix being self-aware (like... what does that even mean in this context?).

Is GPT human-like? Sure... if you stretch the meaning of "human-like" enough (it produces content that is similar to content produced by a human). Is a human GPT-like? That's harder to argue (and I don't see how your argument would support it).

Humans may even "be conceptually mirrors" (whatever that may mean): this still doesn't make mirrors be humans (conceptually or otherwise).

Current language models are much more closer to "Dissociated Press" (i.e. an equation that tells you which which words are more likely, given the context) than to "human thought", and the fact that humans learn by copying does not really change this.

> [...] I'm not sure how much choice they had in this - I suspect NSA/US gov more widely here. [...]

Note that when parent says "you can't trust NIST" and you counter with something along the lines of "that's unfair... NIST acts untrustworthy/knowingly recommends subpar options because of NSA", it doesn't really counter what is being said.

If NIST decisions are based mostly on "whatever the NSA tells them to do", rather than the actual technical merits of the things they recommend, then... yes, they are generally not worthy of trust (blind or otherwise), because you'll always have to double-check their statements against other sources (e.g. your own knowledge, expert cryptographers, etc.).

Fool me once, shame on you; fool me twice, shame on me.

That's the problem of being untrustworthy once in a while... it's easier to lose your reputation than to regain it.

As it is... if you use anything recommended by NIST without first checking with the actual trustworthy community of researchers, you're asking for it.

TL;DR: Trying to justify why the NIST is seen as untrustworthy (or acts as such) does not change the fact that it is seen as untrustworthy by many people (and, as far as I can tell, fairly so).

> it's just not even a choice and I struggle to see how any one would chose a the former as a preferable option.

What if you do not possess a smartphone? What choice do you have, then? Are you just forced into hotel quarantine or do they lend you a smartphone? (Genuine question.)

As it is explained in the "readme" part, in this specific context, "naturally occurring" means that no one has purposefully manipulated any of the images to make them collide: that the images were already published and "out there" and happen to collide. In other words, it does not necessarily imply that the images correspond to natural photographic scenes (which seems to be your interpretation of it).

Besides, you could probably "naturally" obtain such type of colliding images by photographing similar-looking objects against a white (or generally featureless) background. Furthermore, it suggests/demonstrates that similar-looking images with similar backgrounds can lead to unexpected collisions in practice (i.e. "naturally"), even if you do not assume an adversarial scenario.

Are you sure that, if you take a picture of a naked body part, it won't collide with anything that looks similar in their database?

This.

The thing is that some of the techniques commonly applied when training NN are often "good enough" to deal with the presence of corrupted data (e.g. using SGD to optimize a model, while applying weight decay and drop-out, adds a regularization effect that somewhat replicates the effect of assuming errors-in-variables), as long as the input data is not total trash, which deters people from applying more formalized robust approaches to it.

As long as "things kind of work", it is difficult to convince other people to adopt robust methods, particularly due to the existence of a "robustness vs. efficiency" trade-off (which can make robust methods seem additionally "unsexy").

Note: the "multiplex test" is most likely still a PCR test (just 'multiplex PCR' instead of 'single-probe PCR'), so where you say "PCR only detects SARS CoV 2" it should say "the currently-used PCR test only detects SARS CoV 2".

Not at all. Just criticizing his blanket statement, apparently used to justify "doubting the results of US elections": "doubting" just for the sake of doubting, particularly when it feeds into narratives being driven by trolls, disinformation and hidden agendas, doesn't do Democracy or Truth any service.

I'm going to go ahead and assume that the people in question (that are doubting the results of the US elections) are rather "skeptical" towards (so called) mainstream media, but not as much when it comes to the random crap on fox news, ann, youtube and facebook. If this is the case, then this makes them more "useful idiots" than actual "skeptics", in my opinion.

TL;DR: "Doubting blindly" (usually based on whatever Fox News or 4chan is spreading today) is as at least as bad as "trusting blindly".

Ah, yes... the "if I put everything in doubt, even the most painfully obvious things, it makes me seem smart" crowd.

Here's the thing, though: if one is basing their "doubts" on random brainfarts or propaganda pieces they saw on youtube or facebook and decided to trust blindly, they're probably also doing it wrong.

Skepticism that is not accompanied by critical thinking, actual knowledge, intellectual honesty and a dash of humility is useless or worse.

TL;DR: Epistemology is hard, but "just doubt everything" is not it, chief.

Again, which false claim have I made? Be specific and quote me: you need actual examples, not handwaving.

Until you do that, I'm not pursuing this conversation any further. Have a nice day.

EDIT: Also, if you do want to have a conversation, make sure to stick to HN rules and talk about what is being discussed, rather than about me. Thanks.

> This is completely wrong.

Which part? The fact that storing "length" along with a hash is not superfluous?

You can probably find many things which have a SHA hash of "ca978112ca1bbdcafac231b39a23dc4da786eff8147c4e72b9807785afee48bb" (infinite things, if we assume arbitrary-sized inputs), but you can only find ONE thing which has that hash and has length 1. I just made it impossible (not just unlikely) for you to find a collision.

> The known collision attacks for the MD-family and SHA-1 all in fact produce collisions with the exact same length.

Emphasis mine. And note that I did not claim otherwise in my comment.