Category Archives: open science / collab / collective intel

Royen proof of gaussian correlation inequality

hi all theres been a recent shock of awareness of the Royen proof of gaussian correlation inequality, pop-sci publicized by Wolchover for Quanta, a big milestone… this is a nearly ½-century-open problem![a] Quanta funded by Simons institute is one of the top outlets for scientific/ mathematical writing around today. a real community resource/ treasure!

the Royen proof is not exactly my area so cant write a lot on it but do note that its a key case study in dynamics of scientific peer review, and seems like it has some parallels to the ongoing mochizuki proof analysis.[b] it took over ~1½ year for “community” to begin to grasp the correctness of this proof and Wolchover has a nice historical timeline for how others began to notice/ accept it, a mapping of the spread of awareness. it did not help that Royen was somewhat isolated and did not seem to personally contact any cohorts for peer review. he published openly but it got lost in the noise. it shows how community acceptance is sometimes far from a black/ white binary decision, esp for “big problems”.

is there any way to improve peer review? its definitely a bit of an achilles heel of the scientific process. my feeling is that there is no way to improve it very much except maybe to try to increase transparency somehow. its very similar to the problem of “fake news”. how do you measure quality in content? we live in the vast Information Age but as has long been noted, theres a big difference between Information and Wisdom, and in a way peer review is the major mechanism that is designed to separate/ discriminate the two.

Continue reading

AI 2017 — poker milestone passed/ crumbled

hi all. AI technology is really exploding in the last few years. the last big post/ compilation on the subj here was ~½ year ago and the links piled up in a blur since then. the main trigger for this post: the game of poker now seems to have “folded” to computer supremacy. a new paper was published on Deepstack and its highly competitive play, and Libratus is $800K up in a recent match against top experts (top players). my understanding is that there is still some weakness in multiplayer games and that the new breakthru is for 1-1 games, human vs computer, but presumably that razor-thin human edge might also melt away quickly.[a]

poker was a very good game for humans wrt our inherent/ evolved psychology. we (top humans that is) seem to have an intuitive grasp of how to bet based on the strength of cards, including the use of bluffing. it took computers until the 21st century to master this stuff. but it looks like they just passed the threshhold again. in a small surprise, it wasnt done by Deepmind but which is behind many other near-monthly, even verging on weekly breakthroughs.[c]

maybe not by total coincidence, the winning Libratus algorithm involves training a neural network to accurately estimate the search tree, quite similar to the Deepmind Go strategy that made huge headlines just a year ago. the media hasnt picked up on the poker competition as much as it did with Go… is it because cautious/ publicity shy academics have less PR instinct than google? or less budget? but maybe that “relatively low profile” will change in the weeks/ months ahead. hopefully there will be a very high profile contest that again captures widespread public interest/ imagination.

it seems the top poker competitions are typically held in Las Vegas afaik… what would it take to get the computers in that? wouldnt be cool if say Vegas (or some other high profile gambling center) decided to publicize it to attract attn/ tourism? but would the computer algorithms be competitive in the top multiplayer games? there have been increasing/ huge audiences for poker over last few years, not sure what all the factors are in in this surge (internet gambling might play a role…)

its neat to see academia still at the top of competitive research in AI. but that seems to be thinning somewhat over last few years as the massive corporations Google[b], Microsoft, Apple,[g] Facebook, Intel [f] and misc other corps [e] are snapping up AI talent like its a feverish arms race, and to some degree it is. theres also very fast/ dynamic startup/ other merger activity going on, and new research laboratories being founded.[h]

Continue reading

Daniel Sank, Phd/ Google quantum computing lab, stackexchange physics guest speaker Aug 12 2016

physics_chat.png➡ ⭐ ⭐ ⭐ hi all. it is with great pleasure to introduce this next guest and chat series and welcome a larger audience. Daniel Sank, Phd working at Google Martinis quantum computing lab[2] is our 3rd guest speaker[1] in a guest chat series hosted by the physics stackexchange site/ chat room.[13] thanks to moderator David Z, Phd[18] for the suggestion of the idea, coordination and graciously agreeing to hosting in his already well-attended biweekly chat sessions.[11]

we have now hosted two prior events in series with great success with Samuel Lereah, Masters in physics[4][5], and yuggib, Phd math working in mathematical physics, thanks so much to these groundbreaking/ enthusiastic guest speakers for their particularly inspired/ dedicated participation.[5][6]

Continue reading

facebook filtering algorithm/ policy causes big commotion (once again)

❗ 😮 hi all. recently facebook was accused/ implicated in steering news items away from conservative sites. every few years, facebook “faces” these types of PR messes and Zuckerberg is probably nearly a pro at handling them by now. there were a lot of press reverberations.

facebook is like the 800LB gorilla of the media. they embody the “winner takes all” aspect of our economy. most other media outlets are having serious troubles, such as old stalwarts like NYT somewhat “on the ropes”. oh yeah and then there was that crazy gawker judgement recently. hoo, boy!

facebook is rewriting the rules of the media game. much the way that tweaks to googles algorithms cause fat/ huge ripples/ shudders in the mass SEO business (and today, nearly every media outlet is in the SEO business to varying degrees), facebooks small adjustments to their site can have massive repercussions.

its interesting in all this how algorithmic theory is intersecting with media concerns like fairness, impartiality, freedom from censorship, openness, etc… is this a story about algorithms or people? its kind of about the intersection of the two. oh, throw in political pov (right vs left). its a volatile mix. throughout the story, people are grappling with a different landscape. a paradigm shift.

Continue reading

overview/ big trends in cyber collaboration & social networking

hi all, this is a long-in-the-works post of piled up collected/ compiled links on an emerging area that is difficult to name/ circumscribe exactly, but its basically about new forms of collaboration augmented by cyberspace. this is a quickly emerging area that this blog has tracked in some other posts, and heres the latest batch, but also with an eye to identifying and articulating some key larger trends.

the big social network-related “platforms” are Facebook, Twitter, Reddit, Google+, Instagram, StackExchange, Wikipedia. (wikipedia [e] has been covered in depth in prior posts and is typically not regarded as “social networking” but it clearly has strong overlap with the space in general when considered.) they vary quite a bit in nature but there seems to be some convergence going on.

a big area of focus/ adjustment/ contention is what might be called “feed algorithms”. Twitter/ Instagram recently announced changes in their feed algorithms in an attempt to increase “stickiness” aka engagement.[a] Twitter seems to be going thru some wrenching changes lately with large alteration in its executive management and feeling funding pressure/ squeeze. in short it seems to be having trouble monetizing its platform. same executive shakeup went on with Reddit where the toplevel gyrations made major headlines but maybe have now stabilized.[c] in reddits case some of this was due to the intensity of user commitment/ loyalty to the site. reddit has also made changes in feed aka ranking algorithms.

users do not like their feed algorithms altered. unfortunately the big platforms do not allow the user to select the feed algorithm. it seems like a simple/ obvious compromise: allow a few basic feed algorithms, and let users choose their own. but feed algorithms are similar in importance to the “big kahuna” Pagerank and companies want to maximize control over them as “drivers of eyeballs” and influence, and of course wrt advertising. so in a way there is a weird tension verging on “conflict of interest” at the heart of these platforms long noted… in short, users want to use the platforms and control them but the platforms want to control how they are used… software design can steer and sometimes subtly “push back” against user priorities… the software system/ platform can take on its own priorities… aka in the vernacular “taking on a life of its own…”

Continue reading