usgroup

Its nice to see Logica has come on a bit. A year or two ago I tried to use this in production and it was very buggy.

The basic selling point is a compositional query language, so that over-time one may have a library of re-usable components. If anyone really has built such a library I'd love to know more about how it worked out in practice. It isn't obvious to me how those decorators are supposed to compose and abstract on first look.

Its also not immediately obvious to me how complicated your library of SQL has to be for this approach to make sense. Say I had a collection of 100 moderately complex and correlated SQL queries, and I was to refactor them into Logica, in what circumstances would it yield a substantial benefit versus (1) doing nothing, (2) creating views or stored procedures, (3) using DBT / M4 or some other preprocessor for generic abstraction.

I expect he just means deep and foundational models for vision, which is true: they dominate.

I haven't read Hegel before, but I find the abstraction of something like "reason" as separate agent in itself, and then thinking about how its agency manifests a fun way of thinking.

It is, I think, begging the question, with regards to "reasons" autonomy, but it is a mind bending perspective.

From https://www.cambridge.org/core/journals/review-of-politics/a...:

According to Hegel, universal history is the realization of the Idea of Reason in a succession of National Spirits. These are manifested in the deeds of heroes, or world-historical individuals, such as Alexander the Great, Caesar, and Napoleon. However, the Idea of Reason does not work itself out in history in a manner which would seem reasonable on the surface. It is not actualized in this or that of its stages as a consequence of men consciously adopting it as their ideal and striving to translate it into reality through their mode of life and conduct. It is not, as it were, through the imitation of Reason that Reason is realized in history. How, then, does this take place?

Hegel's answer to this question is contained in his doctrine of the Cunning of Reason. Briefly, he holds that history fulfills its ulterior rational designs in an indirect and sly manner. It does so by calling into play the irrational element in human nature, the passions. So Hegel writes: “Two elements, therefore, enter into our investigation: first, the Idea, secondly, the complex of human passions; the one the warp, the other the woof of the vast tapestry of world history.” By “passion” he means those all-consuming emotions of the individual person which are self-regarding in nature. In discussing the question, he writes: “I mean here nothing more than human activity resulting from private interest, from special or, if you will, self-seeking designs—with this qualification: that the whole energy of will and character is devoted to the attainment of one aim and that other interests or possible aims, indeed everything else, is sacrificed to this aim.”

This is hard to do for products when they change over time. Eg if a product is upgraded and repaired, and if it has different versions and so on, it’s no longer a static distribution, and things get complicated.

Shannon refers to two strategies which can be used together. A and B.

Fortran sir, consider Fortran. HPC simulation is exactly what its mostly used for, it runs OpenMP with so little effort, and there's even an Nvidia Fortran optimised for writing GPU kernels.

Fortran is very easy to learn, runs at close to C speed for general code, and it is a very productive language. I personally loved using it.

It is not what Stockfish does though. From the Wiki page:

Stockfish implements an advanced alpha–beta search and uses bitboards. Compared to other engines, it is characterized by its great search depth, due in part to more aggressive pruning and late move reductions.[13] As of September 2024, Stockfish 17 (4-threaded) achieved an Elo rating of 3642 +16 −16 on the CCRL 40/15 benchmark.[14]

See also:

Stockfish historically used only a classical hand-crafted function to evaluate board positions, but with the introduction of the efficiently updatable neural network (NNUE) in August 2020, it adopted a hybrid evaluation system that primarily used the neural network and occasionally relied on the hand-crafted evaluation. In July 2023, Stockfish removed the hand-crafted evaluation and transitioned to a fully neural network-based approach.

https://en.wikipedia.org/wiki/Stockfish_(chess)

Picking a specific definition of a word, expounding its consequences, and then referring to the colloquial usage of the word as a “myth” is just word-play as I said.

Many do think of confounders in an experimental context as just those effects which correlate with both outcome and treatment. The non-sequitur — barring a specific definition — is concluding that since nothing can correlate with random allocation, confounders are impossible by construction.

Why impossible? Because we are talking about the probability of allocation, not the actual allocation, and confounding does not refer to the result. We’d instead say there are imbalanced covariates, but that’s ok because randomisation converts “imbalance into error”. Yet, the covariates may be unknown, and without taking measurements prior to the treatment, how are we supposed to know whether the treatment itself or just membership of the treatment group explains the group differences?

Had we not tested the samples prior to treatment, the result would be what many would call “confounded” by the differences in the samples prior to treatment.

From https://en.wikipedia.org/wiki/Confounding#Decreasing_the_pot..., please note the use of the word:

The best available defense against the possibility of spurious results due to confounding is often to dispense with efforts at stratification and instead conduct a randomized study of a sufficiently large sample taken as a whole, such that all potential confounding variables (known and unknown) will be distributed by chance across all study groups and hence will be uncorrelated with the binary variable for inclusion/exclusion in any group.

You might be thinking about how the evaluation is trained (MCTS) rather than how it’s applied in a chess game?

It doesn't work that way, and it is common to test for known imbalances, or to implement stratified sampling. As another poster said, confounders do not have to be conveniently distributed, even though that is typically assumed. It could be that a big effect is present in a confounder but collected only every N samples because it is sparse. In which case you could have large, randomly allocated but confounded samples.

All these things can and do happen in randomised experiments, but it is still orders of magnitude more interpretable than what can happen in observational studies.

Generally speaking, I've come to expect that from articles with "myths about" in the title outside of the context of the classics or anthropology.

> When we randomise people, these influences will still be operating on the outcome, which will vary across the people randomised to our conditions. Does randomisation mean that all these different effects are balanced somehow? No – not least because confounders do not exist in experimental studies! This is for the simple reason that a confounder is something that affects both the exposure and the outcome, and in an experimental (i.e., randomised) study we test for a difference in our outcome between the two randomised groups.

I do not think this sort of word-play is useful. If your random samples are small (and even if statistically adequate) the chance of confounders in one or both groups of an A/B can be relatively high, even though the selection procedure for treatment is random. So, "No – not least because confounders do not exist in experimental studies!" is misleading if it is expected on the basis that randomisation of treatment allocation somehow makes confounding impossible.

That the possibility of confounding is equally likely in both branches remains true for all sample sizes of an A/B where #A=#B and allocation is random. So, in my opinion, not a myth.

So far as I can tell, the A+B strategy described therein is exactly the principle that Stockfish -- and more or less every other chess engine -- was built on prior to AlphaZero, after which the top engines moved to NN derived evaluation criteria for what is still a pruning tree search.

As others have said, this is par for the course. Competence all around you -- in most places -- is at best a transient phenomena, if you're a top 25% thinker. Assuming you are a top thinker, and you want to be challenged, then you could -- and should -- try to join an elite team elsewhere (e.g. FAANG+), but these opportunities are scarce and not equally available to all of us.

A more general strategy would be to realise what you can do is conditional on the world as it is, and try to make the best of that, improving incrementally as the opportunities arise. But this of course curtails a mind that is trying to produce the best its capacities will allow it.

I think that "making the best of it" is what most of "us" end up doing by the time we are in our 40s, and I would conjecture that here on HN you will find far more beautiful flowers that never blossomed than those that took their full bloom, for much the same reasons you mentioned, and the adaptions given above.

I'd additionally suggest joining the SWI Prolog board. They are a kind and generous community. Jan Wielemaker -- the principal contributor -- often answers questions directly.

Meanwhile, SWI Prolog is a mature and commercially used open-source Prolog distribution. It has a huge number of libraries which will come in useful for real world applications.

Consider starting here: https://book.simply-logical.space/src/simply-logical.html

There isn't much to read; you mostly have to spend your time thinking about implications.

It'll quickly test how into logic you really are :-)

Good luck.

I think it’s a nicely summarised challenge to boot.

It’s doubtless to me that thinking happens without intermediary symbols; but it’s also obvious that I can’t think deeply without the waypoints and context symbols provide. I think it is a common sense opinion.

I think that may be a mostly US bias. I don't think it is true more broadly. E.g. take UK, China, Japan, Russia, France, Germany as an example.

There is a little bit more here:

https://giap.icac.cat/2024/09/09/a-universal-model-for-detec...

For some reason, the article is inaccessible at the time of writing, but it seems to be using a military satellite to map ancient (and presumably shallow) underground water canals for archeological purposes.

Countable infinity does not work like that: two countable infinities are not more than one countable infinity. I think it falls into the "not even wrong" category of statements.

The Wikipedia article is fairly useful: https://en.wikipedia.org/wiki/Countable_set

No breadth first search is still complete given an infinite branching factor (i.e. a node with infinite children). "Completeness" is not about finishing in finite time, it also applies to completing in infinite time.

Breadth first search would visit every node breadth first, so given infinite time, the solution would eventually be visited.

Meanwhile, say a branch had a cycle in it, even given infinite time, a naive depth first search would be trapped there, and the solution would never be found.

Depth first search is not complete if branches can be infinitely deep. Therefore if you're in the wrong infinite branch the search will never finish.

Breadth first search is complete even if the branches are infinitely deep. In the sense that, if there is a solution it will find it eventually.

There are. Tabling (available in most mature implementations) helps when recalculation of the same states is a problem. Meanwhile, custom search strategy is always an option to implement directly in Prolog. You'll see this in many Advent of Code solutions in Prolog when it is applied to path finding puzzles, in which depth first search is rarely a workable solution.

functional logic programming is not equivalent to functional programming.

It ends like this, largely cancelling out the rest of the article:

>In the second iteration of the circle, all of his notes are completely useless, and all of his initial attempts to teach anything fail, because these are different kids with different aptitudes and different interests. Zvonkin, raised in a communist society and a believer in the absolute malleability of human nature, is fairly bowled over by this, especially by how young all these differences are manifesting. Reading between the lines, it sounds like he got quite lucky with his first set of children, and that the second group were much more challenging to teach.8 The most eloquent testimony to this is that after about a year he gives up, and the journal ends abruptly.

Adding covariates to the post analysis can reduce variance. One instance of this is CUPED by there are lots of covariates which are easier to add (eg request type, response latency, day of week, user info, etc).

Tsetlin machines have been around for some time:

https://en.wikipedia.org/wiki/Tsetlin_machine

They are discrete, individually interpretable, and can be configured into complicated architectures.

It doesn't say where the 152 astrologers were sourced, does it? Or how they were qualified? Should astrology be a field with very many impostors, these results would not be unusual.

Generally speaking I suspect it will be difficult to falsify astrology, but more importantly the onus is on the astrologer to prove their ability rather than for others to disprove it.

I think the margin between OpenAI and the next few best competitors is already fairly slim, but as OP mentioned, the margin between the median competitor and the best is also decreasing.

It suggests that playing catch-up progresses quicker than pushing the state of the art.

I don't see how any of that pops a bubble but I think it could transformative for the leading companies (OpenAI, Anthropic, etc) which may become more rank and file service providers.