Hacker News Reader

gone35on Network of fortified towns indicates Amazon was once heavily populated

Or as we describe Mars today, maybe [mutatis mutandis].

gone35on Preparing for Malicious Uses of AI

I am far, far less worried about an assassin's drones using AI to find a politician in a crowd than I am about Facebook using pictures of me that other people have posted and tagged me in, so that my face is used to track my movements, and the movements of every other human on the planet, everywhere we go, and selling that information to everybody who wants a copy, and giving it away at the request of the local police.

Definitely not a lawyer but, as I understand it, depending on jurisdiction and context, some social media postings may be considered "public" information volunteered without "reasonable expectation of privacy" ---in which case, awfully enough, anything goes...

Again, not a lawyer; but I wonder if there should be a right to clear-language, mandatory warnings ---like in cigarrettes--- whenever you are about to post something that will not enjoy "reasonable expectation of privacy" (and hence could be sold or used against you in the future, etc.)..

gone35on Preparing for Malicious Uses of AI

To a sufficiently capable intelligence, encrypted machine code is just as understandable as the most elegant source code.

Depends on what you mean by "sufficiently" and "encrypted"!

I'm not an expert, but my guess is these are still truly difficult, deep unsolved questions... We don't even know yet if one-way functions exist; and existing modes of analysis increasingly seem insufficient to fully account for the apparent recent successes of this last wave of 'AI' on ostensibly non-convex optimization problems! (Not to mention the debates about the possibility of fundamental physical/computational complexity limits to intelligence explosion, and so forth...)

So who knows at this point; but I think it is still an open question whether a future AI could reasonably 'break' even today's commodity encryption.

gone35on Why has CPU frequency ceased to grow? (2014)

Very well put! Not an expert but, as far as I know, parallel OS development definitely seems to be a frequent blind spot in the systems literature...

gone35on The unnecessary demise of Barnes and Noble

[...] I'm a 40 year old man who reads voraciously yet "most" books are neither appropriate, nor interesting to me.

There. Just a friendly suggestion: I think you may be a bit over-extrapolating from your personal experience and your sense of "common sense", to make overgeneralized statements about things as arguably complex and idiosyncratic as "appropriateness" or "interest" in books. (Hence, perhaps, the pushback from other readers.)

gone35on Review of Vivek Wadhwa’s Washington Post Column on Quantum Computing

Scott does come across as petty, and as someone you would never want to work with.

Please don't let his (admittedly sometimes aggressive, unabashed) writing style detract from the man himself. From personal experience, he's unfailingly generous, kind, and amazing to work with.

gone35on My professional opinion as a blockchain researcher: I don’t see the point (yet)

I fail to see the link you are drawing from Sci-Hub to Bitcoin. Care to clarify?

gone35on The Social Dynamics of Programming Together in Dynamicland

Some more background/discussion (after [1]):

[1] http://discuss.tbd.cool/t/dynamicland/86/4

[2] http://vitor.io/on-dynamicland

[3] https://harc.ycr.org/project/realtalk/

[4] https://mobile.twitter.com/redblobgames/status/9072538029311...

gone35on A theory of the learnable

OK, I think I narrowed the 'trick' down. It's actually an interesting existential question.

Correct me if I'm wrong but I think the key step is the use of "positive Boolean semantics"; which, as your Ref. 9 proves, are substantially weaker --and hence, unsurprisingly, far more tractable-- than more conventional "stochastic" or "differential" semantics...

But then Ref. 9 [1] goes on to make, I think, a frankly astonishing, Church-Turing like existential claim in Biology (Sec 3.2, infra):

[...]if a behavior is not possible in the boolean semantics, it is surely not possible in the stochastic semantics whatever the influence forces are.

If that is the case, that would IMO have huge consequences! It would mean, then, that some of the underlying machinery of Biology may turn out to be far simpler than we think: no more pesky self-loops or bistable, mutually inhibitory modules to deal with! Tractable network inference, at last! It would potentially revolutionize computational biology, if true.

But, is it true? I think I see the intuition, but I don't think the case is as clear-cut, with that single "surely" carrying way too much of the rhetorical work... Indeed, the claim hinges on what I think is a rather interesting, non-trivial existential question: informally, if 'something' (of a given type) cannot be denoted in a certain weaker type, does that mean that 'something' cannot exist?

Anyway, not your paper per se; but I think it's an interesting debate nonetheless.

[1] https://hal.archives-ouvertes.fr/hal-01378470

gone35on A theory of the learnable

Interesting work! We definitely need far better inference models in computational biology than what we currently have. I agree with your paper that modeling is unfortunately more of an "art" than a science nowadays... and with huge societal consequences.

Having said that, on a cursory read I think you may be misapplying Valiant's algorithm...

In particular, the original (union bound) PAC guarantee relies crucially on IID samples, so you cannot straightforwardly apply it to time series data and expect the guarantee to hold unchanged. Instead, you should use block bootstrap methods to sample consecutive segments of your time series of a certain size --in which case a (possibly weaker) PAC-like guarantee might hold, provided the dependence across time decays sufficiently fast [1].

I'm also a bit concerned about the semantics of your approach, since I thought gene regulatory inference was/is notoriously intractable, and Valiant's model is very stringent and conservative... So IMO somewhere along the line you are getting a massive free lunch simply by reducing to k-CNF!

Not saying it's wrong per se of course; but I couldn't easily tell exactly where the 'trick' is... So if I were you I would try to communicate more clearly (to dumb non-experts like me) how exactly this particular reduction captures something highly non-trivial in gene regulatory networks to achieve such a (seemingly) drastic speedup..

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4551412/

gone35on Show HN: A Decentralized StatusPage on IPFS

Not a network person by any means, but I think that this points out to what I think is the fatal flaw of IPFS: the apparent "one-wayness" of addresses to content.

Yes, I know that is supposed to be its main feature, but (again, naïvely speaking) it seems to induce a huge, possibly even intractable overhead... Think of, say, programming in C without being allowed to ever rewrite memory contents, or change what a pointer points to!

Again, not an expert; and I'm sure with all the money they raised, people with actual technical expertise in network protocols are hard at work on this and can vouch for the design... But I'm curious anyway how this is not an obvious dealbrealer, given the extreme latency requirements in networking..

gone35on A curated list of fundraising Asian startups

So many mentions of "AI" and "blockchain"; sometimes even in the same phrase... Not a good sign IMO.

gone35on The Magnetic Field Is Shifting. The Poles May Flip. This Could Get Bad

I think it's sometimes OK for potentially existential risks, don't you think?

gone35on Fourier transforms of images (2017)

I know some (rudimentary) harmonic analysis and am familiar with the Shannon-Nyquist theorem but I'm not sure I follow. Care to explain?

Edit: nevermind, I got it [1]. I see, however, the result only holds 'modulo' paraxial waves well-behaved in the linear regime... So it's not that every lens automagically computes a FFT at the speed of light as it seemed at first!

[1] https://en.wikipedia.org/wiki/Fourier_optics#Fourier_transfo...

gone35on Clinical trial finds blood-plasma infusions for Alzheimer’s safe, promising

OT: just wanted to say Fight Aging [1] is an amazing, inspiring model for how modern public interest initiatives should work IMO. Thank you.

[1] https://www.fightaging.org

gone35on Hacking UFO Tech, SETI, and the Boundaries of “Respectable” Science

Buried lede IMO (if factual):

I can say that I have personally spoken to researchers from top universities (Stanford, MIT, Harvard) who have seen the “artifacts” that the article references, and other similar ones that are even more secretive (and perhaps more functional).

gone35on Sudden savant syndrome, in which exceptional abilities emerge after brain injury

10,000 hours

That 10,000 hours thing is largely a Malcolm Gladwell invention, fwiw.

gone35on Physicists Uncover Geometric ‘Theory Space’

Definitely far from clickbait in this particular case. It's hard to tell unless you are somewhat close to the field tho.

gone35on Statistical Computing for Scientists and Engineers

There's a tendency for anyone who learns information theory before statistics to think that information theory is tremendously relevant - but no, rephrasing maximum likelihood or Bayesian methods in information theory terms may sometimes be slightly helpful in thinking about them, but doesn't really add anything fundamental.

And no, there's nothing particularly special or interesting about distributions that maximize entropy subject to some (generally arbitrarily selected) constraint.

Wow, some two heavyweight opinions there. Care to elaborate?

gone35on Neuroscientist Ben Barres has died

It's not just funding.

For one, the dynamics of the known molecular pathways involved are complex [1], and so far non-trivial to manipulate.

Furthermore, even with an actionable mechanism, targeting the tumor itself is highly non-trivial, due to physiology alone. (See e.g. [2]).

[1] http://www.jcancer.org/v07p1497.htm

[2] http://www.nature.com/articles/nrclinonc.2016.119

gone35on Nvidia’s New Policy Limits GeForce Data Center Usage

Japanese original:

https://wirelesswire.jp/2017/12/62658/

Edit:

Apparently it's just that GeForce cards have(/had?) no warranty for use in "data centers" --but academic use is not precluded in of itself:

https://twitter.com/NVIDIAAIJP/status/943141204744585222

https://wirelesswire.jp/2017/12/62667/

Still, the line is somewhat unclear, I wonder how many university/academic cluster admins are aware of the fact..

gone35on Philosophy Needs a New Definition

Paraconsistent logics are demonstrably weaker, so there is that...

gone35on Human Beings Almost Vanished from Earth in 70,000 B.C

Unsurprising. Krulwich is particularly notorious for sloppy science writing/reporting/whatever he does.

gone35on Alan Kay is still waiting for his dream to come true

9:05-12:30 I'm sold. A mindblowing vision for what the universality of computation can truly do... Thank you.

gone35on How to Write Your Own Compiler (2009)

See:

https://en.m.wikipedia.org/wiki/Polytope_model https://www.cs.indiana.edu/~achauhan/Teaching/B629/2010-Fall... http://web.cs.ucla.edu/~pouchet/lectures/doc/888.11.5.pdf

gone35on The Bay Area Is Separating into Red and Green Zones

Yes and no[1]:

Due to the complex history of evolving meanings and contexts, there is no clear or agreed-upon definition of the Third World. Some countries in the Communist Bloc, such as Cuba, were often regarded as "Third World". Because many Third World countries were extremely poor, and non-industrialized, it became a stereotype to refer to poor countries as "third world countries", yet the "Third World" term is also often taken to include newly industrialized countries like Brazil, India and China now more commonly referred to as part of BRIC. Historically, some European countries were non-aligned and a few of these were and are very prosperous, including Ireland, Austria, Sweden, Finland, and Switzerland.

The term Third World is still largely used interchangeably with the least developed countries, the Global South and developing countries.

[1] https://en.m.wikipedia.org/wiki/Third_World

gone35on YC-backed Muzmatch definitely doesn’t want to be Tinder for Muslims

[...]since it is not internally consistent -- and since nothing is -- everything becomes an object of belief at some level.

Giant enthymematic jump right there.

gone35on Want to Be Happy? Buy More Takeout and Hire a Maid, Study Suggests

Edited by Susan T. Fiske, Princeton University. [1]

Evidently so. I wonder what Andrew Gelman will make out of it.

[1] http://www.pnas.org/content/early/2017/07/18/1706541114.full

gone35on Thoughts on OpenAI, reinforcement learning, and killer robots

- Asteroid strike

- Superbug pandemic

- Supervolcano eruption

gone35on Thoughts on OpenAI, reinforcement learning, and killer robots

I support research funding at all levels, and have nothing against mostly theoretical research, but is it really the best use of resources to throw $1 billion at reinforcement learning without any similar investments into addressing mass unemployment and wealth inequality (both of which are well-documented to cause political instability), how existing gender and racial biases are being encoded in our algorithms, and on how to best get this technology into the hands of people working on high impact areas like medicine and agriculture around the world?

Not quite my field, but perhaps such currently intractable, high-impact societal and medical problems do require theoretical breakthroughs after all... I guess I'm just concerned Rachel that --to put it in reinforcement learning terms-- we need both exploration and exploitation.