Profile

Cover photo
Boris Borcic
Attends and plants delegated
Lives in voting to animals
3,582 followers|834,938 views
AboutPostsCollectionsPhotos

Stream

Boris Borcic

Shared publicly  - 
 
 
This is great!
Computer science has for the first time become the most popular major for female students at Stanford University, a hopeful sign for those trying to build up the thin ranks of women in the technology field.
2 comments on original post
2
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
Damned if she does. Damned if she doesn't.

Fuck the Patriarchy.

Artwork by political cartoonist Khalid Albaih
2 comments on original post
18
Josh Hendrix's profile photoJust In Words's profile photoMartin “Jaffa” S's profile photo
4 comments
 
+Just In Words That's true, I hadn't factored in the strong conditioning applied to women, and perpetuated by men. Thanks.
Add a comment...

Boris Borcic

Shared publicly  - 
 
Model promises the silver lining of a 10% rebate on the sea-level rise front.
 
"When Antarctica's air temperature rises, moisture in the atmosphere increases. That should mean more snowfall on the frozen continent. So why hasn't that trend become evident in Antarctica's surface mass balance as climate models predict?
In a new study, scientists used historical records and climate simulations to examine that question. They found that the effect of rising temperatures on snowfall has so far been overshadowed by Antarctica's large natural climate variability, which comes from random, chaotic variations in the polar weather. By mid-century, however, as temperatures continue to rise, the study shows how the effect of human-induced warming on Antarctica's net snow accumulation should emerge above the noise.
The expectation of more snowfall is something of a silver lining as temperatures rise. Global warming is already increasing sea level through melting ice and thermal expansion. The increase in snowfall over Antarctica could help reduce the amount of global sea level rise by 51 to 79 millimeters, or about 2 to 3 inches, by the year 2100, according to the study. That would be a small but important benefit: the Intergovernmental Panel on Climate Change estimates global sea level rise will be at least 10 times that by 2100 under the same high-emissions scenario used in the new study".

(Posted by +rasha kamel​)
When Antarctica's air temperature rises, moisture in the atmosphere increases. That should mean more snowfall on the frozen continent. So why hasn't that trend become evident in Antarctica's surface mass balance as climate ...
View original post
1
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
Art, or Construct?
I bet you've noticed that almost every city has its own creative sculptures or statues, usually tucked along streets or city squares. Some of them are pretty ordinary, but others... Others can be extremely eye-catching and conversation-provoking.
2 comments on original post
8
1
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
"In Lewis Carroll's "Through the Looking-Glass," the Red Queen explains to Alice how a race works in Wonderland, stating, "Now, here, you see, it takes all the running you can do, to keep in the same place." So, too, does this statement hold true in nature. Competitive species are under constant pressure to evolve as rapidly as possible so as to outgun their competition, and this is often referred to as the Red Queen Theory. The rabbit needs to outrun the fox to avoid being killed, whereas the fox needs to catch the rabbit in order to avoid starvation. Well, statistical modeling has also suggested the inverse since 2003: the Red King Theory. If two species are mutualists -- that is, each benefits from the activity of the other -- they should evolve at a slower rate, so as to avoid interrupting their partnership. Makes sense, right? Think again! In a new study published in Nature Communications, comparative genomic analysis shows that the complete opposite may actually be true.
"We originally set out to uncover the genetic basis of mutualistic behavior in ants," said Dr. Benjamin Rubin, recent PhD graduate from University of Chicago and The Field Museum and now postdoctoral researcher at Princeton University. "So, we sequenced the genomes of three mutualistic species of plant-ants and four of their closely-related, non-mutualistic relatives. We were surprised to learn that the mutualists actually had a higher rate of evolution across their genomes than the generalists."
Specifically, the genes that are under significant pressure are those attributed to neurogenesis and muscle activity -- exactly what you might expect to see. Neurogenic genes are tied to behavior, while muscle activation genes likely help the ants protect their host plants through increased activity and speed".
If two species are mutualists -- that is, each benefits from the activity of the other -- the Red King Theory predicts that they should evolve at a slower rate, so as to avoid interrupting their partnership. Makes sense, right? Think again! In a new study published in Nature Communications, comparative genomic analysis shows that the complete opposite may actually be true.
View original post
2
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
We at #artifacia created an evolving list of fundamental papers to learn Deep Learning.

Would like to get your feedback and if we missed out any paper then feel free to comment, we will update the list.

#AI #ML #paper #research
https://medium.com/artifacia/learn-deep-learning-the-hard-way-e5d844f9fbc1#.s3a5a5tyi
There are so many articles about learning Deep Learning but still I decided to write one more. The reason is I find many of those articles…
3 comments on original post
1
Add a comment...
Have him in circles
3,582 people
PetGuide.com's profile photo
John Lawson's profile photo
metrony's profile photo
Lionlio world's profile photo
putra .j's profile photo
S M Moshiur Rahman ABir's profile photo
Afurakan Art Gallery's profile photo
Christian Burger's profile photo
Mar mer's profile photo

Boris Borcic

Shared publicly  - 
 
 
I came across a curious connection between continued fractions and combinatorics on arxiv. It'll take some setup. Let's see how short I can make this...

A "snake graph" is either a "tile" (consisting of 4 vertices and 4 edges) or is built from a snake graph by adding a tile (consisting of 2 new vertices and 3 new edges) above, or to the right of the top right tile in a snake graph. The picture shows an example.

Call adding a tile above a "zig" and a tile to the right a "zag". Think of the first tile as a zig. Then the graph in the picture is constructed from a sequence (zig, zig, zig, zag, zig, zig, zag, zag, zag, zig).

Given a sequence made from two letters (say A and B) you can "compress" it using run length encoding, ie. by listing the length of runs of one letter, starting with the number of A's. Eg. ABBAAABAB becomes (1,2,3,1,1,1).

You can also anti-run length encode it. List the lengths of the anti-runs, i.e. the stretches of alternating letters. So ABBAAABAB, made of anti-runs AB-BA-A-ABAB, becomes (2,2,1,4).

We can anti-run encode the zigs and zags in a snake graph. The pictured graph becomes (1, 1, 3, 5).

Given a sequence a=(a1, ..., an) of integers greater than 0 and with the last greater than 1, we can read it as the anti-run length encoding of zigs and zags and build the corresponding snake graph, G(a).

These sequences also define continued fractions via [a] = [a1, a2, ..., an] = f(a1+1/(a2+1/(...1/an))).

A perfect matching of a graph is a subset of its edges where every vertex in the graph touches precisely one edge in the subset. If an edge between two people expresses compatibility between them, a perfect matching is a way of pairing everyone off as compatible couples. Let m(G) be the number of perfect matchings of G.

Now comes the theorem:
[a1, a2, ..., an] = m(G(a1, a2, ..., an-1))/m(G(a2, a3, ..., an-1)).

In addition, the numerator and denominator are coprime so no cancellation is possible.

I wrote some code to test this theorem, though it wasn't intended to be particularly readable: https://gist.github.com/anonymous/1a18b3b7f5a8e1fac76e4b5dba281c1b

I got this result from a paper on cluster algebras, whatever they are: http://arxiv.org/abs/1608.06568 My notation might not follow the paper exactly.

BTW It's not hard to prove using elementary methods. You can almost read off a proof from my code.

And I probably made at least one mistake above...

6 comments on original post
5
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
Slower Than Light

China recently launched a satellite to test quantum entanglement in space. It’s an interesting experiment that could lead to “hack proof” satellite communication. It’s also led to a flurry of articles claiming that quantum entanglement allows particles to communicate faster than light. Several science bloggers have noted why this is wrong, but it’s worth emphasizing again. Quantum entanglement does not allow faster than light communication.

This particular misconception is grounded in the way quantum theory is typically popularized. Quantum objects can be both particles and waves, They have a wavefunction that describes the probability of certain outcomes, and when you measure the object it “collapses” into a particular particle state. Unfortunately this Copenhagen interpretation of quantum theory glosses over much of the subtlety of quantum behavior, so when it’s applied to entanglement it seems a bit contradictory.

The most popular example of entanglement is known as the Einstein-Podolsky-Rosen (EPR) experiment. Take a system of two objects, such as photons such that their sum has a specific known outcome. Usually this is presented as their polarization or spin, such that the total must be zero. If one photon is measured to be in a +1 state, the other must be in a -1 state. Since the outcome of one photon affects the outcome of the other, the two are said to be entangled. Under the Copenhagen view, if the entangled photons are separated by a great distance (in principle, even light years apart) when you measure the state of one photon you immediately know the state of the other. In order for the wavefunction to collapse instantly the two particles must communicate faster than light, right? A popular counter-argument is that while the wavefunction does collapse faster than light (that is, it’s nonlocal) it can’t be used to send messages faster than light because the outcome is statistical. If we’re light years apart, we each know the other’s outcome for entangled pairs of photons, but the outcome of each entangled pair is random (what with quantum uncertainty and all), and we can’t force our photon to have a particular outcome.

The reality is more subtle, and vastly more interesting. Although quantum systems are often viewed as fragile things where the slightest interaction will cause them to collapse into a particular state, that isn’t the case. Entangled systems can actually be manipulated in a variety of ways, and you can even manipulate them to have a specific outcome. I could, for example, create pairs of entangled photons in different particular quantum states. One state could represent a 1, and the other a 0. All my distant colleague needs to do is determine which quantum state a particular pair is in. But to do this my colleague would need to make lots of copies of a quantum state, then make measurements of these copies in order to determine statistically the state of the original. But it turns out you can’t make a copy of a quantum system without knowing the state of the quantum system. This is known as the no-cloning theorem, and it means entangled systems can’t transmit messages faster than light.

Which brings us back to the experiment China just launched. The no cloning theorem means an entangled system can be used to send encrypted messages. Although our entangled photons can’t transmit messages, their random outcomes are correlated, so a partner and I can use a series of entangled photons to generate a random string we can use for encryption. Since we each know the other’s outcome, we both know the same random string. To crack our encryption, someone would need to make a copy of our entangled states, which can’t be done. There are ways to partially copy the quantum state, which would still improve the odds of breaking the encryption, but a perfect copy is impossible.

So entanglement doesn’t give us faster than light communication, but it may make it a bit easier to keep our secrets secret.

Quantum entanglement won't let us communicate faster than light, but it might help us keep our secrets.
41 comments on original post
2
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
They’ve officially been classified as “critically endangered.” Make sure the plight of the orangutans doesn’t go ignored - add your name and help publicise this crisis >>
View original post
3
Add a comment...

Boris Borcic

Shared publicly  - 
 
 
We share pics we didn't take
to amuse people we don't care about
in a social network that we hate.

Comment stolen from Facebook
11 comments on original post
4
Add a comment...
Boris's Collections
People
Have him in circles
3,582 people
PetGuide.com's profile photo
John Lawson's profile photo
metrony's profile photo
Lionlio world's profile photo
putra .j's profile photo
S M Moshiur Rahman ABir's profile photo
Afurakan Art Gallery's profile photo
Christian Burger's profile photo
Mar mer's profile photo
Education
  • and plants delegated
    present
Basic Information
Gender
Male
Story
Tagline
in the beginning, animals voted
Introduction
Born a twin and somewhat later, first distracted of girls by computers with ram capacity in the kilobytes range; that is, truly parsimonious computers.

Former Taglines

Dark energy is relic pollution from warp drives! Solve two cosmic mysteries - the Fermi Paradox and Dark Energy - by each other. Transpose the AGW issue to the Universe, to defuse the illusion we can turn our back on the problem by moving to exoplanets. Make fun of Dark Energy etc having inspired the possibility of warp drives, by assuming cosmic history has already articulated both together, but in reverse order.

Does God Sexist? - The misogyny of many religious texts and traditions, entangles the matter of affirming God with that of affirming misogyny.

Ambiguities are like microbes: the pathogens steal attention! - most people will equate microbes with the tiny fraction of pathogens, and will ignore in particular the helpful microbes without which we wouldn't even survive. I claim a similar situation applies to ambiguities, even more remote from awareness.
Work
Occupation
with their genitals
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
voting to animals
Previously
Geneva
Links