Sunday, December 11, 2011

Various approaches to constructing AI

Self-Improving Artificial Intelligence




Whole Brain Emulation: The Logical Endpoint of Neuroinformatics?



Saturday, December 10, 2011

Time estimation ability predicts mathematical intelligence

Being good at estimating time can be a useful skill on its own, but it may also indicate higher mathematical intelligence as well, according to a new study published in the Dec. 7 issue of the online journal PLoS ONE.

A test of 202 students, evenly divided between males and females, revealed that those subjects who were better at estimating the durations of a series of short tones were also more likely to correctly answer various mathematical questions relative to their more poorly estimating counterparts.

This correlation was not seen with a general intelligence test, suggesting that time estimation is specifically related to mathematical intelligence.

The authors, led by Peter Kramer of the University of Padua in Italy, conclude that this relationship is likely due to a common reliance on spatial ability. "Encouraging this tendency might help improve mathematical intelligence and satisfy one of modern society's greatest needs", says Dr. Kramer.


Source EurekaAlert!

Swarms of bees could unlock secrets to human brains

Scientists at the University of Sheffield believe decision making mechanisms in the human brain could mirror how swarms of bees choose new nest sites.

Striking similarities have been found in decision making systems between humans and insects in the past but now researchers believe that bees could teach us about how our brains work.
Experts say the insects even appear to have solved indecision, an often paralysing thought process in humans, with scouts who seek out any honeybees advertising rival nest sites and butt against them with their heads while producing shrill beeping sounds.

Dr James Marshall, of the University of Sheffield's Department of Computer Science, who led the UK involvement in the project and has also previously worked on similarities between how brains and insect colonies make decisions, said: "Up to now we've been asking if honeybee colonies might work in the same way as brains; now the new mathematical modelling we've done makes me think we should be asking whether our brains might work like honeybee colonies.

"Many people know about the waggle dance that honeybees use to direct hive mates to rich flower patches and new nest sites. Our research published in the journal Science (on December 9), shows that this isn't the only way that honeybees communicate with each other when they are choosing a new nest site; they also disrupt the waggle dances of bees that are advertising alternative sites."

Biologists from Cornell University, New York, University of California Riverside and the University of Bristol set up two nest boxes for a homeless honeybee swarm to choose between and recorded how bees that visited each box interacted with bees from the rival box. They found that bees that visited one site, which were marked with pink paint, tended to inhibit the dances of bees advertising the other site, which were marked with yellow paint, and vice versa.

Tom Seeley of Cornell University, author of the best-selling book Honeybee Democracy said "We were amazed to discover that the bees from one nest box would seek out bees performing waggle dances for the other nest box and butt against them with their heads while simultaneously producing shrill beeping sounds. We call this rough treatment the 'stop signal' because most bees that receive this signal will cease dancing a few seconds later."

Dr Patrick Hogan of the University of Sheffield, who constructed the mathematical model of the bees, added: "The bees target their stop signal only at rivals within the colony, preventing the colony as a whole from becoming deadlocked with indecision when choosing a new home. This remarkable behaviour emerges naturally from the very simple interactions observed between the individual bees in the colony."

Monday, December 5, 2011

Discovery of the fastest-rotating massive star ever recorded

(Santa Barbara, Calif.) –– An international team of scientists has found the fastest-rotating massive star ever recorded. The star spins around its axis at the speed of 600 kilometers per second at the equator, a rotational velocity so high that the star is nearly tearing apart due to centrifugal forces. This confirms a prediction put forward by astrophysicist Matteo Cantiello, a postdoctoral fellow with UC Santa Barbara's Kavli Institute for Theoretical Physics, who contributed to the discovery published this week in theAstrophysical Journal Letters.

This is an artist's concept of the fastest-rotating massive star found to date. The massive, bright young star, called VFTS 102, rotates at about two million kilometers per hour. Centrifugal force from this dizzying spin rate has flattened the star into an oblate shape, and spun off a disk of hot plasma, seen edge on in this view from a hypothetical planet. The star may have "spun up" by accreting material from a binary companion star. Scientists believe that the rapidly evolving companion star later exploded as a supernova. The whirling star lies 160,000 light years away in the Large Magellanic Cloud, a satellite galaxy of the Milky Way.

The observations were made at the European Southern Observatory's Very Large Telescope, at the Paranal Observatory in Chile, as part of a survey of the heaviest and brightest stars in a region called the Tarantula Nebula. The Tarantula Nebula is a region of star formation located in a neighboring galaxy called the Large Magellanic Cloud, about 160,000 light years from Earth. The reported star, VFTS 102, is extremely hot and luminous, shining about 100,000 times more brightly than the sun. According to the research team, this star had a violent past and was ejected from a double star system by its exploding companion star.

Cantiello and collaborators explained that stars could reach such rapid rotation via a "cosmic dance" with another star so close that gravity strips gas from its surface. "This gas falls onto the companion star, increasing the mass and spinning it up," said Cantiello. "Similar to a tennis ball spinning fast after being hit by a glancing blow, a star rotates quickly after being hit off-center by the in-falling gas."
Cantiello previously predicted the possibility of observing this type of star. He reported this theoretical finding with Sung-Chul Yoon, Norbert Langer, and Mario Livio in a paper published in 2007 in Astronomy & Astrophysics Letters. This theoretical investigation of stars in binary systems predicted extreme rotational velocities after mass accretion. The observed rotational velocity for the star agrees with this prediction.


The star is unusual not only because it rotates so fast, but also because it is moving away from its neighboring stars at a velocity of about 70,000 miles per hour, or 30 kilometers per second. "Having been part of a binary system could explain this space oddity," said Cantiello. "It has been known for over 40 years that a star in a massive binary system can be shot away from its surroundings when the companion ends its life in a spectacular explosion called a supernova. In our theoretical calculations we noticed that the 'spun-up' star would also be moving from its surroundings at a high rate. It is very exciting to find a star that matches both of these predictions."

The star is located close to a pulsar and a supernova remnant, which may be left over from the companion star that once spun-up the observed star. If confirmed, this would provide additional support for the theoretical explanation put forward by Cantiello and collaborators in 2007.
Cantiello said that this star may produce dramatic fireworks as it dies. Such a rapidly rotating, massive star is believed to be the progenitor of some of the brightest explosions in the universe: gamma-ray bursts. These occur when the star's fast rotation produces powerful jets of light and matter.

Smallest habitable world around sun-like star found

Astronomers have found the smallest planet ever detected in the habitable zone around a star like the sun.
The new planet was found with the KeplerMovie Camera telescope, which searches for signs that a star's light has dimmed because a planet has passed between it and the telescope – an event called a transit.
Kepler-22b is just 2.4 times as wide as Earth(Image: NASA/Ames/JPL-Caltech)
"This discovery supports the growing belief that we live in a universe crowded with life," team member Alan Boss of the Carnegie Institution for Science said in a statement. "Kepler is on the verge of determining the actual abundance of habitable, Earth-like planets in our galaxy."
The planet, named Kepler-22b, lies 600 light years away around a star of the same type (called G) as the sun. It is about 2.4 times as wide as Earth and orbits its star every 290 days, right in the middle of its star's habitable zone, where liquid water can exist on an object's surface.
Transit observations cannot pinpoint its mass, however. Astronomers have used other telescopes to search for signs that the planet's gravitational tugs are causing its host star to wobble, but so far have not detected any wobbles. That means the planet's mass must be less than 36 times that of the Earth.
It is close in size to a class of planets called super-Earths, which are up to about 2 times as wide as Earth. "We have no planet like this in our solar system," says Bill Borucki, Kepler's chief scientist at NASA's Ames Research Center in Moffett Field, California. He announced the find on Monday at the Kepler Science Conference at NASA Ames.

Just right

The allowed mass range means the planet could be rocky and could contain water, Borucki says. Ground-based observations in mid-2012, when the patch of sky where the planet lies is more easily visible, could help astronomers nail down the planet's mass. That will help them identify its composition.
Two previous rocky planet candidatesMovie Camera have been found in the habitable zones of their stars, but in both cases the stars were cooler than the sun.
And neither candidate was found right in the middle of its star's "Goldilocks" zone, which could boast the best conditions for hosting life as we know it. Kepler-22b's surface is probably a balmy 22 °C, Borucki said.

Scanning for ET

The Kepler telescope has been staring at more than 150,000 stars between the constellations Cygnus and Lyra for the past 1000 days. The Kepler team has now found more than 2300 candidate exoplanets, about 1000 more than itreported in February. Ten of those span no more than about twice Earth's width.
To confirm a new planet, scientists must observe three of its transits. Mission scientists saw the first transit of Kepler-22b three days after Kepler began collecting data in 2009. The third transit appeared in December 2010. "It's a great gift," Borucki said. "We consider this our Christmas planet."
"It's conceivable that these new planet candidates and their [potential] moons could have life," Borucki said.
The SETI Institute in Mountain View, California, will observe the new candidates with its Allen Telescope Array of radio telescopes in California in the hopes of detecting signals from any extraterrestrial civilisations there, said the institute's Jill Tarter. The array had been offline since April due to budget cuts but restarted observations on Monday after raising funds by partnering with the US air force and crowdsourcing donations.

Saturday, December 3, 2011

Entangled diamonds blur quantum-classical divide


Two diamonds as wide as earring studs have been made to share the spooky quantum state known as entanglement. The feat, performed at room temperature, blurs the divide between the classical and quantum worlds, since typically the quantum link has been made with much smaller particles at low temperatures.
One laser pulse entangled two diamonds and the next measured the entanglement
Entanglement is one of the weird aspects of quantum mechanics, where the fates of two or more particles are intertwined – even when they are physically far apart. Electrons, for example, have been entangled, so that changing the quantum spin of one affects the spins of its entangled partners.
Macroscopic objects, on the other hand, are supposed to mind their own business – flipping one coin shouldn't force a neighbouring flipped coin to come up heads.
But that's just what happened with two 3-millimetre-wide diamonds on a lab bench at the University of Oxford. Physicists there led by Ka Chung Lee andMichael Sprague were able to show that the diamonds shared one vibrational state between them.
Other researchers had previously shown quantum effects in a supercooled 0.06-millimetre-long strip of metal, which was set in a state where it wasvibrating and not vibrating at the same time. But quantum effects are fragile. The more atoms an object contains, the more they jostle each other about, destroying the delicate links of entanglement.

Fleeting link

Cooling an object down to fractions of a degree above absolute zero was thought to be the only way to keep atoms from doing violence to each other.
"In our case we said, let's not bother doing that," says Ian Walmsley of Oxford, head of the lab where the diamonds were entangled. "It turns out all you need to do is look on a very short timescale, before all that jostling and mugging around has a chance to destroy the coherence."
The team placed two diamonds in front of an ultrafast laser, which zapped them with a pulse of light that lasted 100 femtoseconds (or 10-13 seconds).
Every so often, according to the classical physics that describes large objects, one of those photons should set the atoms in one of the diamonds vibrating. That vibration saps some energy from the photon. The less energetic photon would then move on to a detector, and each diamond would be left either vibrating or not vibrating.
But if the diamonds behaved as quantum mechanical objects, they would share one vibrational mode between them. It would be as if both diamonds were both vibrating and not vibrating at the same time. "Quantum mechanics says it's not either/or, it's both/and," Walmsley says. "It's that both/and we've been trying to prove."

Same state

To show that the diamonds were truly entangled, the researchers hit them with a second laser pulse just 350 femtoseconds after the first. The second pulse picked up the energy the first pulse left behind, and reached the detector as an extra-energetic photon.
If the system were classical, the second photon should pick up extra energy only half the time – only if it happened to hit the diamond where the energy was deposited in the first place. But in 200 trillion trials, the team found that the second photon picked up extra energy every time. That means the energy was not localised in one diamond or the other, but that they shared the same vibrational state.
Entangled diamonds could some day find uses in quantum computers, which could use entanglement to carry out many calculations at once.
"To actually realise such a device is still a way off in the future, but conceptually that's feasible," Walmsley says. He notes that the diamonds were entangled for only 7000 femtoseconds, which is not long enough for practical applications.

Quantum limit

The real value of the experiment may be in probing the boundary between quantum mechanics and classical physics. "We think that it is the first time that a room-temperature, solid-state system has been demonstrably put in this entangled quantum state," Walmsley says. "This is an interesting avenue for thinking about how quantum mechanics can emerge into the classical world."
Erika Andersson of Heriot-Watt University in Edinburgh, UK, agrees.
"We want to push and see how far quantum mechanics goes," she says. "The reported work is a major step in trying to push quantum mechanics to its limits, in the sense of showing that larger and larger physical systems can behave according to the 'strange' predictions of quantum mechanics."