As the “Brains Vs. Artificial Intelligence: Upping the Ante” poker competition nears its halfway point, Carnegie Mellon University’s AI program, called Libratus, is opening a lead over its human opponents — four of the world’s best professional poker players. Libratus had amassed a lead of $459,154 in chips in the 49,240 hands played by the end of Day Nine. (from cmu.edu)
One of the pros, Jimmy Chou, said he and his colleagues initially underestimated Libratus, but have come to regard it as one tough player. “The bot gets better and better every day,” Chou said. “It’s like a tougher version of us.”
Brains Vs. AI, which began Jan. 11 2017 at Rivers Casino in Pittsburgh, pits Chou and three other leading players — Dong Kim, Jason Les and Daniel McAulay — against Libratus in a 20-day contest in which they will play 120,000 hands of Heads-Up, No-Limit Texas Hold’em poker. All four pros specialize in this two-player, unlimited bid form of Texas Hold’em and are considered among the world’s top players of the game.
While the pros are fighting for humanity’s pride — and shares of a $200,000 prize purse — Carnegie Mellon researchers are hoping their computer program will establish a new benchmark for artificial intelligence by besting some of the world’s most talented players.
Libratus was developed by Tuomas Sandholm, professor of computer science, and his Ph.D. student, Noam Brown. Libratus is being used in this contest to play poker, an imperfect information game that requires the AI to bluff and correctly interpret misleading information to win. Ultimately programs like Libratus also could be used to negotiate business deals, set military strategy or plan a course of medical treatment — all cases that involve complicated decisions based on imperfect information.
In the first Brains Vs. AI contest in 2015, four leading pros amassed more chips than the AI, called Claudico. But Sandholm said he’s feeling good about Libratus’ chances as the competition proceeds. “The algorithms are performing great. They’re better at solving strategy ahead of time, better at driving strategy during play and better at improving strategy on the fly,” Sandholm said.
Chou said he and the other pros have shared notes and tips each day, looking for weaknesses they can each exploit.
“The first couple of days, we had high hopes,” Chou said. “But every time we find a weakness, it learns from us and the weakness disappears the next day.”
The change from day to day is not unexpected, Sandholm said. Each night after poker play ends, the Pittsburgh Supercomputing Center’s Bridges computer performs computations to sharpen the AI’s strategy. During the day’s game play, Bridges is used to compute end-game strategies for each hand.
“The people at the Pittsburgh Supercomputing Center have done a great job,” Sandholm said, noting the staff has moved workloads around to enable the computer to be used in the competition. Since the beginning of the contest, the center has increased the number of Bridges’ computer nodes assigned to the poker tournament.
Cosmic rays may have just unveiled a hidden chamber within Egypt’s most famous pyramid. An international team led by Kunihiro Morishima at Nagoya University in Japan used muons, the high-energy particles generated when cosmic rays collide with our atmosphere, to explore inside Egypt’s Great Pyramid without moving a stone. (from newscientist.com)
Muons can penetrate deep into rock, and get absorbed at different rates depending on the density of the rock they encounter. By placing muon detectors within and around the pyramid, the team could see how much material the particles passed through.
“If there is more mass, fewer muons get to that detector,” says Christopher Morris at Los Alamos National Laboratory, who uses similar techniques to image the internal structure of nuclear reactors. “When there is less mass, more muons get to the detector.”
By looking at the number of muons that arrived at different locations within the pyramid and the angle at which they were travelling, Morishima and his team mapped out cavities within the ancient structure.
This type of exploration – muon radiography – is perfect for sensitive historical sites as it uses naturally occurring radiation and causes no damage to the structure.
The team mapped the pyramid’s three known chambers – the subterranean chamber, the Queen’s chamber, and the King’s chamber – along with connecting corridors. They also detected a new large void above the Grand Gallery that connects the King and Queen’s chamber. This new void is approximately the same volume as the Grand Gallery. The team believes it’s another oversized tunnel similar in dimensions to the Grand Gallery that is at least 30 metres long.
The team used three different muon detectors, starting with nuclear emulsion film within the Queen’s chamber. Like photographic film is exposed to light to make a photo, the emulsion reacts to muons and makes a record of their paths.
Once their initial findings indicated a potential cavity, they confirmed it by placing an instrument that emits a flash of light when struck by muons within the pyramid. Outside the pyramid, they also used detectors that record muons indirectly when the high-energy particles ionise the gas inside. After several months in position to record muons, all three methods confirmed a void in the same location.
“It’s marvelous,” Morris says, noting that the long exposure times increase the robustness of the results. “What they’ve seen is fairly definitive,” he says, although it will take drilling and cameras to determine if the cavity is a structural chamber, or a void created by a long-forgotten collapse.
A team led by Luis Alvarez first tried using muon radiography to map pyramids in 1970, but they were unable to detect new voids. If confirmed, this would be the first newly rediscovered chamber within the Great Pyramid in more than a century.
“I’d love to be there when they first stick a camera through a drill hole,” Morris admitted. “It’s not every day we discover a chamber in a pyramid.”
Journal reference: Nature, DOI: 10.1038/nature24647
Lasers are everywhere nowadays: Doctors use them to correct eyesight, cashiers to scan your groceries, and quantum scientist to control qubits in the future quantum computer. For most applications, the current bulky, energy-inefficient lasers are fine, but quantum scientist work at extremely low temperatures and on very small scales. For over 40 years, they have been searching for efficient and precise microwave lasers that will not disturb the very cold environment in which quantum technology works. (from phys.org)
A team of researchers led by Leo Kouwenhoven at TU Delft has demonstrated an on-chip microwave laser based on a fundamental property of superconductivity, the ac Josephson effect. They embedded a small section of an interrupted superconductor, a Josephson junction, in a carefully engineered on-chip cavity. Such a device opens the door to many applications in which microwave radiation with minimal dissipation is key, for example in controlling qubits in a scalable quantum computer.
The scientists have published their work in Science on the 3rd of March.
Lasers have the unique ability to emit perfectly synchronized, coherent light. This means that the linewidth (corresponding to the color) is very narrow. Typically lasers are made from a large number of emitters (atoms, molecules, or semiconducting carriers) inside a cavity. These conventional lasers are often inefficient, and dissipate a lot of heat while lasing. This makes them difficult to operate in cryogenic environments, such as what is required for operating a quantum computer.
Superconducting Josephson junction
In 1911, the Dutch physicist Heike Kamerlingh Onnes discovered that some materials transition to a superconducting state at very low temperatures, allowing electrical current to flow without any loss of energy. One of the most important applications of superconductivity is the Josephson effect: if a very short barrier interrupts a piece of superconductor, the electrical carriers tunnel through this non-superconducting material by the laws of quantum mechanics. Moreover, they do so at a very characteristic frequency, which can be varied by an externally applied DC voltage. The Josephson junction is therefore a perfect voltage to light (frequency) converter.
Josephson junction laser
The scientists at QuTech coupled such a single Josephson junction to a high-quality factor superconducting micro-cavity, no bigger than an ant. The Josephson junction acts like a single atom, while the cavity can be seen as two mirrors for microwave light. When a small DC voltage is applied to this Josephson junction, it emits microwave photons that are on resonance with the cavity frequency. The photons bounce back and forth between two superconducting mirrors, and force the Josephson junction to emit more photons synchronized with the photons in the cavity. By cooling the device down to ultra-low temperatures (< 1 Kelvin) and applying a small DC voltage to the Josephson junction, the researchers observe a coherent beam of microwave photons emitted at the output of the cavity. Because the on-chip laser is made entirely from superconductors, it is very energy efficient and more stable than previously demonstrated semiconductor-based lasers. It uses less than a picoWatt of power to run, more than 100 billion times less than a light globe.
Low-loss quantum control
Efficient sources of high quality coherent microwave light are essential in all current designs of the future quantum computer. Microwave bursts are used to read out and transfer information, correct errors and access and control the individual quantum components. While current microwave sources are expensive and inefficient, the Josephson junction laser created at QuTech is energy efficient and offers an on-chip solution that is easy to control and modify. The group is extending their design to use tunable Josephson junctions made from nanowires to allow for microwave burst for fast control of multiple quantum components. In the future, such a device may be able to generate so-called “amplitude-squeezed” light with has smaller intensity fluctuations compared to conventional lasers, this is essential in most quantum communication protocols. This work marks an important step towards the control of large quantum systems for quantum computing.
Using a state-of-the-art device for measuring mass, researchers at the National Institute of Standards and Technology (NIST) have made their most precise determination yet of Planck’s constant, an important value in science that will help to redefine the kilogram, the official unit of mass in the SI, or international system of units. (from nist.gov )
The new NIST measurement of Planck’s constant is 6.626069934 x 10−34 kg∙m2/s, with an uncertainty of only 13 parts per billion. NIST’s previous measurement, published in 2016, had an uncertainty of 34 parts per billion.
The kilogram is currently defined in terms of the mass of a platinum-iridium artifact stored in France. Scientists want to replace this physical artifact with a more reproducible definition for the kilogram that is based on fundamental constants of nature.
Planck’s constant enables researchers to relate mass to electromagnetic energy. To measure Planck’s constant, NIST uses an instrument known as the Kibble balance, originally called the watt balance. Physicists widely adopted the new name last year to honor the late British physicist Bryan Kibble, who invented the technique more than 40 years ago.
NIST’s Kibble balance uses electromagnetic forces to balance a kilogram mass. The electromagnetic forces are provided by a coil of wire sandwiched between two permanent magnets. The Kibble balance has two modes of operation. In one mode, an electrical current goes through the coil, generating a magnetic field that interacts with the permanent magnetic field and creates an upward force to balance the kilogram mass. In the other mode, the coil is lifted at a constant velocity. This upward motion induces a voltage in the coil that is proportional to the strength of the magnetic field. By measuring the current, the voltage and the coil’s velocity, researchers can calculate the Planck constant, which is proportional to the amount of electromagnetic energy needed to balance a mass.
There are three major reasons for the improvement in the new measurements, said physicist Stephan Schlamminger, leader of the NIST effort.
First, the researchers have much more data. The new result uses 16 months’ worth of measurements, from December 2015 to April 2017. The increase in experimental statistics greatly reduced the uncertainty in their Planck value.
Second, the researchers tested for variations in the magnetic field during both modes of operation and discovered they had been overestimating the impact the coil’s magnetic field was having on the permanent magnetic field. Their subsequent adjustment in their new measurements both increased their value of Planck’s constant and reduced the uncertainty in their measurement.
Finally, the researchers studied in great detail how the velocity of the moving coil affected the voltage. “We varied the speed that we moved the coil through the magnetic field, from 0.5 to 2 millimeters per second,” explained Darine Haddad, lead author of the NIST results.
In a magnetic field, the coil acts like an electric circuit consisting of a capacitor (a circuit element that stores electric charge), a resistor (an element that dissipates electrical energy) and an inductor (an element that stores electrical energy). In a moving coil, these circuit-like elements generate an electrical voltage that changes over time, said Schlamminger. The researchers measured this time-dependent voltage change to account for this effect and reduced the uncertainty in their value.
This new NIST measurement joins a group of other new Planck’s constant measurements from around the world. Another Kibble balance measurement, from the National Research Council of Canada, has an uncertainty of just 9.1 parts per billion. Two other new measurements use the alternative Avogadro technique, which involves counting the number of atoms in a pure silicon sphere.
The new measurements have such low uncertainty that they exceed the international requirements for redefining the kilogram in terms of Planck’s constant.
“There needed to be three experiments with uncertainties below 50 parts per billion, and one below 20 parts per billion,” Schlamminger said. “But we have three below 20 parts per billion.”
All of these new values of the Planck’s constant do not overlap, “but overall they’re in amazingly good agreement,” Schlamminger said, “especially considering that researchers are measuring it with two completely different methods.” These values will be submitted to a group known as CODATA ahead of a July 1 deadline. CODATA will consider all of these measurements in setting a new value for Planck’s constant. The kilogram is slated for redefinition in November 2018, along with other units in the SI.
The SESAME project has reached an important milestone: the first complete cell of this accelerator for the Middle East has been assembled and successfully tested at CERN. (from home.cern/about/updates/2015/04/sesame-passes-important-milestone-cern)
SESAME is a synchrotron light source under construction in Jordan. It will allow researchers from the region to investigate the properties of innovative materials, biological processes and cultural artefacts. SESAME is a unique joint venture that brings together scientists from its Members: Bahrain, Cyprus, Egypt, Iran, Israel, Jordan, Pakistan, the Palestinian Authority and Turkey. Not only is SESAME an important scientific project, it is also helping to build bridges between diverse cultures in a part of the world that usually hits the headlines for its conflicts.
CERN has been a strong partner to SESAME, providing technical expertise for the design and procurement of accelerator components. In particular, CERN is responsible for the magnets of the SESAME storage ring and their powering scheme, under a project largely funded by the European Commission (FP7 CESSAMag).
Within this project, CERN has been collaborating with SESAME to design, test and characterize the components of the magnetic system, which is now in production. The main contracts have been split among different companies in Cyprus, France, Israel, Italy, Spain, Switzerland, Turkey and the UK, with additional in-kind support (material and personnel) from Iran, Pakistan and Turkey.
The test carried out at CERN together with colleagues from SESAME aimed at assembling a full periodic cell of the machine, one of the 16 which make up the regular structure of the ring. Besides the magnets themselves, this involved also the girder support structure as well as the vacuum chamber for the beam.
“We already knew that the various individual elements fulfil and even exceed the specifications,” says Attilio Milanese, the CERN engineer in charge of the magnets, who is well satisfied since “this test now confirms that all the subsystems work harmoniously together”.
The magnet production is now in full swing. After acceptance tests, these components will be shipped in batches to SESAME by the end of the year, where installation and commissioning of the main synchrotron is planned for 2016.