HN user
Coincidentally I recently watched these videos about this topic. Both great channels!
3D Guide - How to Build the Perfect Medieval Castle https://youtu.be/Syjg6PHYFBo?si=JceRfeOks3hOVqWu
How to Lay Siege to a Medieval Fortress (1000-1300) https://www.youtube.com/watch?v=aQ7hTNoK-OA&t=1900s
An app to estimate the risk of your job being automated and how to hedge against it professionally and financially
You can look at one of the discs here: https://sanctuaryonthemoon.com/discs
HyDE is the way to go! Just ask the model to generate a bunch of hypothetical answers to the question in different formats and do similarity on those.
Or even better, as the OP suggests standardise the format of the chunks and generate a hypothetical answer in the same format.
It may interest you that Guy Podjarny, one of the Snyk founders, now has an AI coding company (https://www.tessl.io/about) that looks like a competitor of yours
There seems to be a deliberate, implicit value judgement about "busybodies" that would explain the negative connotations:
Bassett hypothesizes that “in countries that have more structures of oppression or patriarchal forces, there may be a constraining of knowledge production that pushes people more toward this hyperfocus.”
Great post! I've been looking to get into the guts of large scale model training (I'm half-way between the design and application layer of LLMs, mostly in python, sometimes a bit of c++) and this will be a great reference to have.
PS. appreciate it if anyone can recommend more material like this
I think the issue of test set contamination is important, but it’s academic - when a model contains a good enough distilled representation of arguably all the code out there, does it really matter whether it can generalise OOD?
Realistically how many of the practical use cases where it’ll be applied will be OOD? If you can take GPT4 there then you are either a genius or working on something extremely novel so why use GPT4 in the first place?
I understand the goal is for LLMs to get there, but the majority of practical applications just don’t need that.
Its the outcome when agents don't co-operate
Cooperation is a choice, as you very well say immediately below the above statement.
Haha fair enough but I think we digress. I did get the warning when I clicked.
The point is I see no evidence why HTTP is better for viewability than HTTPS. I showed there are cases where it’s not. You showed there are cases where it’s irrelevant. But the point remains… when is it worse and is it enough to sacrifice the security benefits?
Ah I see, sorry to not address that - got carried away on entropic musings. It seems that is not an assumption he’s making, but it’s what he’s proven:
when a group of atoms is driven by an external source of energy (…) and surrounded by a heat bath (…), it will often gradually restructure itself in order to dissipate increasingly more energy.
To be honest I cannot go any deeper without reading the paper.
Agree. I cannot comment on how probable or how widespread, but only that it is possible, and in fact it is arguably a certainty based on the premises above. Now if entropy was the only driver then self-replicating systems would be the norm. But alas, the Fermi paradox. I think we are asking too much.
Could it be falsified? I think if you could show that natural selection led to the survival of less dissipative life forms then you could claim falsification.
How I understand it:
Entropy (according to Boltzmann) is proportional to the number of microstates that can give rise to a particular macrostate of a system. The macrostate with the highest number of possible microstates is the uniform one, where all the accessible microstates are equally likely. So if by the 2nd law entropy must increase, the system will tend towards the uniform configuration.
In other words, if a force is pushing water, the configuration with the highest number of states is that where all the particles of water are also moving i.e. indistinguishable microstates, doesn't matter how you rearrange the moleculles, it will look the same as they are all moving uniformly. It would be extremely unlikely that there were a pocket of water that is mysteriously still in the middle of the stream no? Clearly there are not a lot of rearrangements of the mollecules that will keep the macrostate the same, so low entropy, which tends not to be the case.
(Edit) How this connects to the "desire to dissipate energy efficiently":
Essentially, when we talk about "energy" we really mean "free energy". This is the amount of work that we can extract from a system. This is nothing but a measure of how far a current system is to it's maximum entropy state. So dissipating energy = increasing entropy.
The mindblowing part for me is the connection that the process of extracting free energy is the same process that moves a system to its most likely state, uniformity, high entropy. So somehow, the ability to _accelerate_ an already inevitable process lets us reconfigure other systems _away_ from their most likely state!! So if we imagine the arrow of time to progress at the average rate of entropic decay, we are essentially reversing it for some systems by accelerating it for others!!!
Man...
Indeed these are normative claims, not descriptive. The emergence of life is is an extremly complex stochastic process, and likely impossible to describe. However, it does seem plausible that if a) high dissipative systems are more likely to perdure in time given the 2nd law, b) self-replication enhances the global entropic dissipation of a category of systems (because there are more) and c) the process of developing such a system is easier done by replication than by the same evolutionary pathway that led to the appearance of said system in the first place, then given enough time a self-replicating system MUST appear. I would like to see a formalisation of c) though, my formulation is quite tenuous.
I don't see why it would have to lead to a non-Darwinian explanation of evolution. As has been said by in the parent comment, survivability might just be a high-level manifestation of entropic drive of self-replicating systems. It's remarkable that this is a claim that can be falsified, whereas Darwinian evolution, well..
Quoting England:
I am just saying that from the perspective of the physics, you might call Darwinian evolution a special case of a more general phenomenon.
I think this is exactly alligned with your Newtonian-GR analogy.
Why would that be the case? Every modern browser actively discourages visiting HTTP sites so I imagine a non-negligible amount of users decide not to visit the page.
I find too many foundational problems and contradictions within e/acc to take it seriously. They are speculating with ideas that IMO have potential but the priciples are more a poorly though out sci-fi ideological statement than a philosophically comprehensive theory. Just the first few points of the manifesto make no sense to me:
The overarching goal for humanity is to preserve the light of consciousness.
If all physical processes are subservient to a thermodynamical "telos", how could any manifestation of free will emerge? Not even going to get into how problematic it is to cite conciousess as a fundamental goal of when we cannot even assess its existence in other beings.
Technology and market forces (technocapital) are accelerating in their power and abilities. This force cannot be stopped.
What? Cannot be stoped? How could humanity possibly play any role in an unstoppable process? Surely no one can deny capitalism to be the result of specific choices in human history? Hence why we can concieve alternatives to capitalism and can very well choose them over capitalism.
Technocapital can usher in the next evolution of consciousness, creating unthinkable next-generation lifeforms and silicon-based awareness.
Can? How could it not if the claim is that it's unstoppable? Concisouness again?
HUMANS HAVE AGENCY RIGHT NOW. WE CAN AFFECT THE ADVENT OF THE INFLECTION IN THIS PROCESS
Scratches head...
I would love to have a rigorous debate about some of these ideas but come on... I was recently in an X space with some of the writers of this manifesto and was very disappointed.
Fair enough, that doesn't sound great.
That being said:
Devices likely to be affected by the certificate expiry are those that don’t get updated regularly, like embedded systems that are designed not to automatically update or smartphones running years-old software releases. Users running older versions of macOS 2016 and Windows XP (with Service Pack 3) are likely to face issues, along with clients dependent on OpenSSL 1.0.2 or earlier, and older PlayStations that haven’t been upgraded to newer firmware.
Still not convinced we can claim HTTPS should not be used in the general, or in this particular, case.
And a follow up from a few years later: https://www.quantamagazine.org/first-support-for-a-physics-t...
So what you're saying is:
HTTPS is secure -> a bug made 2.6% of all certificates less secure -> that's why we shouldn't use HTTPS?
I'm quite confused how that holds logically. You can either argue HTTPS is fundamentally not secure or is irrelevant; or regret that there was a bug but support using HTTPS
https://www.troyhunt.com/heres-why-your-static-website-needs...
The whole video is great, but particularly this: https://www.youtube.com/watch?v=_BNIkw4Ao9w&t=1063s
https://doesmysiteneedhttps.com/
There are a lot of arguments about why, even non-secret data connections need encryption. The first and most obvious is that not doing so immediately puts a target on the encrypted data.
By no means an expert but to me it seems like an oversight not to.
Nice stuff. I love the idea of really learning electronics, but I find it very much beyond what I would need to build the things I want to build.
PS. Guys HTTPS please!
IMO learning is at the service of building. I learn what I need to build. If GPT can do the first 30% of any project, then I can focus my learning on the remaining 70%. Just think of GPT as another layer of abstraction.
Sure, pure learning is enjoyable but also ephemeral. Building/doing = learning, yes, but no need to do everyhting.
Nice work!
How does it compare against FAISS?
I am curious why isn't the network always in an "interference" state, but sometimes it collapses into a "restricted" ONB?
the neural networks we observe in practice are in some sense noisily simulating larger, highly sparse networks
This seems somewhat related to a point made by Ilya Sutskever here [1]: NNs can be though of an approximation to the Kolmogorov compressor. Speculating, one could say any network is a projection of the ideal compressor (which argualy perfectly represents all n-features in an n-dimensional ONB) into a lower dimentional space, hence the interference. But why is there not always such an interference?
Honestly accelerators are overrated IMO (been through one). It's a forcing function for you to build 24/7 + an opportunity to network.
99% of the advice you can find online. Start with the YC library.
Dexter Chua's notes merit a mention IMO for supporting material/summary notes: https://dec41.user.srcf.net/notes/