peppery

Since you did the hard work of parsing rich metadata already, it would be even cooler if your network visualization oriented nodes by some of this information. Here the 'hiveplot' idea (https://hiveplot.com/ ) is often even more useful than e.g. springloaded or UMAP based layouts; clustering into semantically-meaningful categories into axes (say, city or arrondissement? years open? cuisine? an explicit phylogeny from oldest culinary grandparents to youngest?) then choosing a coordinate to localize nodes on the axes (total node degree? prix? "les plus" tags?...) automatically compels us think about salient features of the data.

Agreed, idiosyncratic voice is so life- and mind- affirming in papers. (Do you mind sharing examples of three papers that you did enjoy slowly and change your conceptual life?)

For those in (or soon traveling to) Southern California, the LA County Museum of Art is now featuring a lovely computer art exhibit that includes physical outputs from Harold Cohen, among other works from algorithmic luminaries of art: https://www.lacma.org/art/exhibition/coded-art-enters-comput...

"When art critics get together they talk about Form and Structure and Meaning. When artists get together they talk about where you can buy cheap turpentine." — Picasso.

^Culture is good at romanticizing the "dreamer" as divorced from (& higher than) the "doer"/implementer. Picasso might protest. This post is helpful in inviting us examine this instinct/tradition.

But that good contrariness doesn't excuse us from being more thoughtful about the chain of deduction underlying the titular claim. First, many big ideas—Maxwell's equations of electromagnetism; that software should be Free as in both Beer and Software; that unruly Democracy could possibly be sometimes a (messily, weirdly) good way of organizing people—might naturally require more doers/thinkers to implement than just the mind (few minds?) who happened to crystallize it. Non-unitary stoichiometries for progress are the rule, not the exception. Nudging a culture is already a many-body problem, & because details matter, and details scale exponentially with levels of abstraction, a project's success can improve with the number of engineering minds adding leverage to advance it.

If we take the above earnestly—that making ideas useful usually requires more people than who happened to express an idea—then noticing that more job postings exist for implementory/engineering roles than for "science" roles actually says nearly nothing about whether we as a technological culture are out of balance with science vs engineering, up to how poorly we know about the typical ratio of implementers versus dramers.

It could be that there are plenty of good "scientific" ideas in circulation; maybe what separates us from progress is earnest implementation, reflected by empirical over-demand for engineers (as this post seems to mainly argue). The aggressive scaling laws for improving AI (along existing paradigms but broader compute) are tempting support for this conclusion.

But personally, I think it comes down to your position on this underlying question. Do you believe that fundamentally _better_ data paradigms--eg those that actually compute differently & more (super)humanly--will come from ideas already articulated in the conceptual universe? Or do you think that the key to smarter data science, if it exists, has yet to be invented and may little resemble the ideas dominantly in circulation?

If the latter, then we may most desperately need data _scientists_, in addition to engineers! In the sense that society would totally benefit from generating 1 new idea from “science,” even if 1000+ had been funded but not panned out. This is true on a global, pro-social sense, but also likely on an individual basis: if you are a thinker than surely partly what matters is what role could maximize your ∂impact/∂effort, and probabilistically, science of data is at least competitive with engineering if you think the future will look different than the present and needs to be invented.

Tao is likely inviting us (just as many physical/probabilistic laws do) to view any arbitrary function as relatively "thicker"/"fuzzier" than an infinitely-thin, infinitely-tall spike function at a certain value: the Dirac delta function (https://en.wikipedia.org/wiki/Dirac_delta_function). If you convolve ≡ integrate this Dirac delta function (located at some value x) against any function g(t), by construction the integral is zero everywhere except at t=x, so the result is an infinitely thin slice of g at x, exactly g(x) (the 'sifting property,' https://math.stackexchange.com/questions/1015498/convolution...). Now imagine you begin to thicken/fuzz the spike; now you begin to accumulate the behavior of g(t) not just exactly at x, but also at points nearby, getting a schmeared representation of g. Deforming our spike to an arbitrary function of interest in this way gives an arbitrary convolution schmear.

While possible, the notion that asymptomatic[†] cases are due to a separate less-pathogenic strain is both indeed a possibility researchers are thinking about, and yet also not (a priori) the only or perhaps most probable explanation for the wide variance in clinical outcomes we see. Other important, perhaps dominant, factors include heterogeneity in people's immune responses to virus--common in other conditions--and (possibly) a dose-dependence (e.g. if you are exposed to a high viral load e.g. by intense or prolonged exposure, some reports (but far too few for definiteness yet) are that clinical outcomes may be poorer). Though there are different COVID strains in circulation (see the amazing data tracking of https://nextstrain.org/ncov), with regards to the proposed hypothesis: there is no evidence that these strains show any difference in virulence (see e.g. the perspective of Francois Balloux at UCL: https://twitter.com/BallouxFrancois/status/12395362423558225...). Many groups are attempting to scale up environmental genomic testing for COVID (see again the nextstrain site).

[†] Note that currently, researchers are rather vigorously debating the true proportion of asymptomatic cases (or distinguishing them from pre-symptomatic cases)--we need more widespread e.g. antibody-based testing to answer this more confidently than we can by indirectly fitting coarse time-series to simplified models.

Lest others also experience HN's hug-of-death of the original source, Google's cached version (https://webcache.googleusercontent.com/search?q=cache:bkjo_-...) works as interim alternative version of the original PDF before the link returns or e.g. Wayback Machine has a chance to archive it.

(This aside, heed @asdfasgasdgasdg's very prudent note of caution; other independent and more reputable replications are essential.

As of this writing there are at least 19 such hydroxychloroquine trials in US (https://clinicaltrials.gov/ct2/results?term=Hydroxychloroqui...) and more beyond (though WHO clinical trial site listing them is also apparently under critically-heavy load).)

This article presumes the premise that "what you love"/"your passion" is the same as "what you can love"/the set of all "passions" you have not yet discovered. This is untrue.

Especially for young people, the amount of time that you have been alive is small compared to your lifetime. What you currently know to be interesting is correspondingly a small subset of the number of things you can find interesting over a lifetime, and an even smaller subset of the things which you could find fulfilling to work on with many lifetimes. (For those who have lived longer, your life experience makes it even more likely you can identify fulfilling connections/facets of the universe to study.)

The challenge is to find the intersection between what you can be riveted to work on, and what society values (in whatever its flawed wisdom) or can be invited to value. This is not trivial, but the statistics of the universe are on your side.

What sort of society would we be if e.g. Nikola Tesla/Jame Clark Maxwell/Mozart/etc. had followed this advice?

To aspire is human, powerful, fulfilling. To eat is practical. It is possible to do both. Society needs people who persist in that pursuit.

This is a richly intriguing phenomenon, with interesting implications!

For those interested in reading more, this article (on stochastic resonance's potential importance in biological sensing) is edifying: https://www.physik.uni-augsburg.de/theo1/hanggi/Papers/282.p... (Hänggi, Peter. "Stochastic resonance in biology: how noise can enhance detection of weak signals and help improve biological information processing." ChemPhysChem 3.3 (2002): 285-290.)

(Perhaps this is an example of how biological systems can value accurate sensation so highly that they invent ingenious sensing schemes to achieve high performance.)

Thanks for your sharing your thoughts!

> One missing ingredient is a probabilistic model of the thing you're imaging

Ooh, this (making the prior for the true image signal more informative by incorporating knowledge of the structure of the signal) is clever. Here, when you say model, you mean a description that is based on the (bio)physics of your sample? (E.g., knowing that objects being imaged obey diffusion equations informs your maximum likelihood estimation of the true signal?)

> Secret because I have a dream for a crazy startup based on this idea which I don't have the means to do now. Although I generally hate being secretive about, like, knowledge and for sure the value creation happens during execution - but just humor me this time, ok?

Of course; such is the right (and joy) of an innovator to define how one's own idea is disseminated/actualized! (~: It is intriguing that you feel your idea has the character of best being pursued via a startup (addressing some crucial unmet commercial need), rather than via the academic model (e.g. transformative Nature publication) more commonly used for improved microscopy techniques. Best of luck in this pursuit; I look forward to seeing your startup's innovations someday soon!

> You need to understand the whole picture, how it works on each level, and how the levels fit together - then you can run thought experiments. A good generative model of the world.

This idea of developing a "good generative model of the world" is a beautiful aspiration for all of us to have. Thanks for your insights!

Your work with ML and images sounds incredible!

Do you have a link (e.g. on arxiv or elsewhere) that describes your approach using ML for image reconstruction in greater detail? How would you recommend building up one's combined intuition in optical theory, the relevant ML techniques, and the biological substances themselves, to the level where you can innovate in this task as you have done?

Also, for those interested in the concept of building better images using higher fidelity simulations of the microscope itself, presumably Andrew meant studies along these lines: https://arxiv.org/abs/1702.07336

Woohoo, thank you!

The probability that an average entrepreneur succeeds might indeed be assessed fairly by observing that only one entrepreneur in a hundred ever reaches the 1% of wealth or impact.

But this assessment itself seems to suffer from a bias of another sort--the notion that entrepreneurial risk is uniformly distributed over the population of entrepreneurs. Actually, there will be some startups whose risk will be much lower than that of the "average" emerging company (due to the fact that their venture idea satisfies some market need, discovered either by genuine insight/ingenuity or by luck).

It is the belief--however appropriate--that one's own venture falls within this enlightened category of diminished risk that propels founders to pursue their ventures in the face of such an aggregate track record.

It was the American physicist Albert Michelson (the creator of the beautiful Michelson interferometer design which enabled very precise measurements of the speed of light) who said "it seems probable that most of the grand underlying principles have been firmly established . . . An eminent physicist remarked that the future truths of physical science are to be looked for in the sixth place of decimals."

Subsequently, the sentiment of this statement has been misattributed to Lord Kelvin.

Instead of short selling an equity (which is indeed complicated not so much in its difficulty but in its potentially unlimited downside), you can buy a put option. For the cost of the option you receive the right (not the obligation) to sell a stock at a specified price. If the stock price then goes down below that that price, simply buy the stock at the lower market price then use your put option to sell it at the higher price. The downside is limited to the cost of the option.

It is a good idea! Another excellent editable wiki for proofs has existed for some time: https://www.proofwiki.org/wiki/Main_Page It is nice to see growth in this fertile space.

This project sounds very interesting! I'd also love to hear more about any details you make available.

Thanks for the reference to the wavelet transform--it is nice to consider a concrete context where this operation can be employed.