Profile cover photo
Profile photo
Ryan Reece
Science is a way of getting informed about reality.
Science is a way of getting informed about reality.

Ryan's posts

Post has shared content
// Bert Dreyfus passed a few days ago. I don't see any announcements online, but his close friends and colleagues are mourning on FB.

I started working on AI as a grad student in the final days on the AI Winter, circa 2005. The explosion of machine learning and the Yudkowsky/Bostrom style AGI arguments still lay in the future ( Most AI work was being done in niche technical corners of computer and cognitive science, where there was very little contact with mainstream philosophy. For the beginning grad, it seemed like the philosophical discussion had been exhausted in the 70's and 80's, at the height of Winter, in the arguments from Searle and Dreyfus.

Searle's arguments are terrifically bad, and grounded in nothing more than Searle's own stubborn intuitions. Unfortunately these arguments remain very popular, despite having almost no practical implications. Fortunately, Searle's own career seems to be over now, due to numerous sexual scandals. Hopefully he takes his bad arguments with him.

Dreyfus' arguments were far more subtle, and were deeply grounded in the traditions of existential philosophy and phenomenology. "What computers (still) can't do" made enough cogent arguments to have a palpable impact on the technical development of computer science and robotics. Dreyfus' arguments deserve some credit for the shift of focus to "embodied cognition" that has characterized both research and theory in the decades since. Dreyfus paved the way for roboticists like Rodney Brooks and Hans Morevec.

Dreyfus made his share of bad arguments too. In the first edition to "On the Internet" (published 2001), Dreyfus used his embodiment argument to conclude that searching the internet would be practically impossible. The argument is obviously mistaken, and the second edition of the book (published 2008) removes it entirely. I don't think this mistake is accidental; I think it speaks directly to the limitations of Dreyfus-style arguments that reigned during the AI Winter. I also don't think it's coincidental that these retractions become necessary at the dawning of our new golden age of AI. Diving deep into Dreyfus' views is worth a longer discussion for a different time.

The point is that these discussions are still worth having today. Dreyfus' contributions to the AI literature are important and profound. He's been a steadfast critic of AI, and he's made the AI community better for it. We'd all do well to pay attention.

I had lunch with Dreyfus once as a very young grad student, at a conference at Notre Dame. The conference was on philosophical anthropology (= not AI), and I was at a table with other philosophical legends (Charles Taylor, etc). But I wanted to talk with Dreyfus about whether or not Deep Blue could play chess. And he entertained me, and we had a very pleasant conversation. From what I remember, he offered mostly Socratic questions to help me articulate my thoughts; I don't remember taking away any profound insight from the conversation. But I did get the clear impression that he was a kind, patient, and good-natured person and teacher. I've never heard anyone say otherwise.

image via Taylor Carmen

Post has attachment
My new hobby: transcribing Kool A.D. raps because you cannot find a lot of his lyrics online. He put out 10 recordings last year (some are many hours long) --- seriously most prolific rapper of 2016.

(The ?? indicate words I'm not 100% about. Help.)

Peace to the Standing Rock Sioux
Clean water for Flint Michigan
Free Mumia
Free Palestine
Free Leonard Peltier
Free all political prisoners
Legalize all drugs, dude
Free everybody
Abolish jail
Institute a basic income
Existence is a human right, basically at this point right, in our technical evolution?
I got I call, hold up, yo
Listen to your guy’s wild elocution, ooh, son
Rooftop gardens, solar energy
I think I’m Ajamu Baraka
Bustin’ tags on rigged voting machines
Peace alive, destroy, uh, police property
Build a new concept of well being out of mutually understood concepts of the beautiful
It’s really quite simple
Manchurian candidate style
Hella various aisles to walk down
I’m with the diamond or the center lane
Sometimes I’m sitting in a black leather chair moving faster than the earth’s rotation
Keep my notation
A beast to no nation
At least I believe to be
Every single moment of life is both precious and earned
Every single knowledge is learned
At the instant of air touching lungs
The spirit is the hum, the tributary
From which all rivers flow
Metaphor is never specific
It never means to be
Study my intrinsically executed style
Styles upon impeccable styles
Impeccatuity?? Who is he?
I rapped over the hook, but I don’t give a fuck, Kool A.D.
I wanna tell the world, turn up, it’s your party
Ha ha, hardy har har
I’m mighty funny
But I’m no dummy
Rest in peace, Prince Nelson
Peace to Gene Wilder, Ali, Bomaye
Peace alive, Richard Pryor, bench one??
Austin Prior??
Fake thugs is liars
Peace alive, Tupac
Jah Rastafari
Everybody grindin’
Trust your own struggle, man
Follow your central vibrations
Imbibe libations
It’s a vivation
Be your own private satellite space station
Your spirit stayed tethered to the forever infinite untethered
Never nothing total everything relevant
To anything ever, under, or over understood
Whatever your boy told you good, ha

Post has attachment

Post has shared content
Let us read what we paid for

Imagine a business like this: you get highly trained experts to give you their research for free... and then you sell it back to them.  Of course these experts need equipment, and they need to earn a living... so you get taxpayers to foot the bill.  

And if the taxpayers want to actually read the papers they paid for?   Then you charge them a big fee!

It's not surprising that with this business model, big publishers are getting rich while libraries go broke.  Reed-Elsevier has a 37% profit margin!

But people are starting to fight back — from governments to energetic students like ‎Alexandra Elbakyan here.

On Friday, the Competitiveness Council —a gathering of European ministers of science, innovation, trade, and industry—said that all publicly funded scientific papers published in Europe should be made free to access by 2020

This will start a big fight, and it may take longer than 2020.   But Alexandra Elbakyan isn't waiting around.

In 2011, as a computer science grad student in Kazakhstan, she got sick of paying big fees to read science papers.  She set up SciHub, a pirate website that steals papers from the publishers and sets them free.

SciHub now has 51,000,000 papers in its database.  In October 2015, Elsevier sued them.  In November, their domain name was shut down.  But they popped up somewhere else.  By February, people were downloading 200,000 papers per day.   Even scientists with paid access to the publisher's databases are starting to use SciHub, because it's easier to use.

Clearly piracy is the not the ultimate solution. Elbakyan now lives in an undisclosed location, to avoid being extradited.  But she gave the world a much-needed kick in the butt.   The old business model of get smart people to work for free and sell the product back to them is on its way out.

For more, read:

John Bohannon, Who's downloading pirated papers? Everyone, Science, 28 April 2016,

and especially the SciHub Twitter feed:

Also read this:

Martin Enserink, In dramatic statement, European leaders call for ‘immediate’ open access to all scientific papers by 2020, Science,
27 May 2016,

The Dutch government is really pushing this!  Congratulations to them!


Post has shared content
Why do theoretical physicists travel a lot?

Find out:  


Image credit Sophia Bennett/CERN ©CERN – for terms of use see:

Post has shared content
> A recurring theme in natural philosophy is the tension between the God’s-eye view of reality comprehended as a whole and the ant’s-eye view of human consciousness, which senses a succession of events in time. Since the days of Isaac Newton, the ant’s-eye view has dominated fundamental physics. We divide our description of the world into dynamical laws that, paradoxically, exist outside of time, and initial conditions on which those laws act. The dynamical laws do not determine which initial conditions describe reality. That division has been enormously useful and successful pragmatically, but it leaves us far short of a full scientific account of the world as we know it. The account it gives—things are what they are because they were what they were—raises the question, Why were things that way and not any other?

The God’s-eye view seems, in the light of relativity theory, to be far more natural. Relativity teaches us to consider spacetime as an organic whole whose different aspects are related by symmetries that are awkward to express if we insist on carving experience into time slices.

// This whole essay is wonderful, but the paragraph below makes me want to cheer out loud.

> The work of designing algorithms can be considered as a special form of teaching, aimed at extremely clever but literal-minded and inexperienced students—that is, computers—who cannot deal with vagueness. At present those students are poorly motivated and incurious, but those faults are curable. Within 100 years they will become the colleagues and ultimately the successors of their human teachers, with a distinctive style of thought adapted to their talents.

via +Jon Lawhead

Post has attachment
Professor Luciano Floridi: "Ethics in the Age of Information"

Post has attachment
The Democrats Abroad Global Presidential Primary allows Americans abroad to vote and be represented by "global" delegates independent of those of the states. Sanders won 9 delegates total, while Clinton got 5.

I think it's more interesting to note that Sanders won a majority of the votes in every Country/Region except for three: Dominican Republic, Nigeria, and Singapore (Nigeria only 4 votes cast). The rest went for Bernie, often with large margins.,_2016

Post has shared content
judgement day

Post has shared content
It's been on my todo-list for a while now to understand the issues of "radius of convergence" in perturbative QFT. I'm still puzzled by what the issue is since the partial sums of perturbation series are so predictive and work so well empirically in the Standard Model. Really grateful to John Baez for discussing this topic in this series of posts.
The trouble with QED

If you're trying to understand charged particles and radiation in a way that takes special relativity and quantum mechanics into account, you need QED.

That stands for quantum electrodynamics. Feynman, Schwinger and Tomonaga invented this theory - with lots of help - around 1948. In QED we often compute answers to physics problems as power series in the fine structure constant

α ≈ 1/137.036

This number says how strong the electric force is. For example, if you have an electron orbiting a proton, on average it's moving about 1/137.036 times the speed of light.

We can compute lots of things using QED. A great example is the magnetic field produced by an electron. The electron is a charged spinning particle, so it has a magnetic field in addition to its electric field. How strong is this magnetic field?

With a truly heroic computation, physicists have used QED to compute this quantity up to order α⁵. This required computing and adding up over 13,000 integrals. If we also take other Standard Model effects into account we get agreement with experiment to roughly one part in a trillion!

This is often called the most accurate prediction of science. However, if we continue adding up terms in this power series, there is no guarantee that the answer converges. Indeed, in 1952 Freeman Dyson gave a heuristic argument that makes physicists expect that the series diverges, along with most other power series in QED!

I explain that argument in this blog article. I'm especially happy because I think I've made it a bit more precise. But it's not a proof: just an argument that something very strange must happen if the answer converges.

Currently, the consensus among physicists is that ultimately QED is inconsistent. I explain why. But again, there's no proof. We need some mathematicians to help settle these questions!
Wait while more posts are being loaded