Month: June 2015

Photos, videos, PDF documents and location data: the permissions requested by some apps give them access to more information than users are aware of. EPFL researchers have come up with a tool to better follow and manage these risks.

Exactly what personal data are we putting on the web in the era of the cloud? Often much more than we want or even imagine. That is what researchers at EPFL’s Distributed Information Systems Laboratory (LSIR) have discovered. They have developed a tool, PrivySeal, that informs users exactly what data they are agreeing to share when they accept the permissions of various apps available on the internet.
Popular websites like Google Drive, Dropbox and OneDrive generally have strict and longstanding privacy rules. The only catch is that these websites are often used as a platform for related apps that provide cloud-based services like photo editing, PDF merging and editing, content searching and analysis, to name just a few.
“We want to let users know the risks they take when using these services and give them a better way to assess these risks,” said Hamza Harkous, a PhD candidate at the LSIR and the lead researcher on this project.

Where, when and with whom?

The “Accept” that users mechanically click so they can use an app to touch up a photo, for example, can open the door to the users’ entire photo collection – and to the sometimes very detailed date- and location-related information tagged to the photos. Someone could use that to figure out where and with whom the user had a drink on Tuesday at 6pm or where the user went on vacation last summer. Blindly accepting the permissions of a document filing service can also provide access to other types of data. And as a result, a third party could see who the user’s main collaborators were and what topics were discussed, or figure out the user’s opinions on various issues by analysing the most frequently occurring words.
“If they want to use these services, users sometimes have no other choice but to provide access to additional information, and in so doing they sacrifice some of their privacy,” said Rameez Rahman, an LSIR researcher who oversaw the project. “The requested permissions often cover a much broader array of data than the services really require to function.”
Prepare to be surprised!
To assess the scope of the problem, the researchers analysed more than seventy apps that are offered on two cloud platforms: Google Drive and Dropbox. The results showed that nearly half of them had this type of privacy problem.
The website developed by the LSIR researchers is now publicly accessible at privyseal.epfl.ch. It provides clear, step-by-step guidance allowing users to figure out exactly what they have authorised their apps to do and access. The personalised results are then displayed in graphical format, including different-sized circles showing the people, places and companies with which the users have the most contact on a day-to-day basis. The website also provides specific details on the consequences of various types of permission requests. After learning what data their apps can access, users can see which of the data are really needed for the apps to function properly and which are not.

References:http://phys.org/

Prior to the mid-18th century, it was tough to be a sailor — you couldn’t set out for a specific destination and have any real hope of finding it quickly if the trip required east-west travel.

At the time, sailors had no reliable method for measuring longitude, the coordinates that measure how far east and west one is from the international dateline. Longitude’s key was accurate timekeeping, as the English watchmaker John Harrison knew, and clocks just weren’t accurate yet.

To measure distance, measure time

“If you want to measure distances well, you really need an accurate clock,” said Clayton Simien, an NSF-funded physicist at the University of Alabama-Birmingham. His current research on cutting-edge atomic clocks was inspired, while he was an undergraduate, by Dava Sobel’s book “Longitude: The True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time” (Walker & Co., 2001).

By the 1700s, sailors had figured out they could measure latitude by studying the sun and its location at various times of the day, so north-south travel was not so problematic. However, the place where longitude equals zero, known as the International Date Line, does not have a basis in nature. As evidenced by several relocations of the prime meridian, located in Greenwich, England since 1884, its placement is arbitrary. After all, who’s to say whose daybreak starts the Earth’s next rotation? [Atomic Clock Is So Precise It Won’t Lose a Second for 15 Billion Years ]

“How you define time is pretty much arbitrary in the sense that in the past we defined a year by using how long it takes the earth to rotate around the sun,” Simien said. “So, basically, any periodic, consistent motion can be the basis for a clock. I used to joke with my relatives that I can say that time is how long it takes me to walk up and down five flights of stairs, while eating a bag of Doritos. But that wouldn’t be a good definition of time. Some days I might be tired, so I move slower. You wouldn’t want to base time on something that can vary so much.”

Sailors figured out that as they traveled east, time moved ahead — the sun set earlier than expected, for example. In fact, based on current parameters for time, for every 15 degrees of longitude a person moves east, the local time moves ahead an hour. That meant longitude could be grossly measured by contrasting the time of day from two places: a ship’s location and its departure port. But, like climbing stairs while eating chips, such measurements also require standards, which for those sailors meant building a clock from materials that didn’t rust and didn’t swell or contract with heat and cold, preserving a reference for the time “back home.”

Harrison, that English watchmaker, put together a clock of wooden wheels — replacing the prior state-of-the-art, a pendulum, with something called a grasshopper escapement, which on its first voyage in 1736 helped identify a 60-mile course divergence for his ship. As a result, he won the Longitude Prize for building the first compact marine chronometer.

The quest to improve timekeeping continues today, as scientists look at new materials that are even more rugged and precise, eliminating variables that might distort accurate timekeeping.

Atomic clocks in GPS satellites work with ground-based clocks so that positioning signals are synchronized as much as possible. Atmospheric distortions present challenges that can limit signal accuracy beyond the most precise atomic clock’s scope. So, while the U.S. Air Force operates the more than 30 GPS satellites in orbit, several government agencies, including NSF, the U.S. National Institute of Standards and Technology, the U.S. Department of Defense, and the U.S. Navy are invested in atomic clock research and technology.

But today’s research isn’t just about building a more accurate timepiece. It’s about foundational science that has other ramifications.

One second equals one ‘Mississippi’ or ~9 billion atom oscillations

Atomic clocks precisely measure the ticks of atoms, the back and forth transition between two different atomic states. The atoms, commonly cesium, can transfer from the ground state to an excited state, but only if the frequency is just right. The trick to this process is finding just the right frequency to move directly between the two states and overcoming errors, such as Doppler shifts, that distort rhythm.

Today’s most accurate atomic clocks use laser-beam photons to “cool” atoms to low temperatures, to within a millionth of a degree of absolute zero. This reduces Doppler shifts and provides a long time to observe the atoms, which improves an atomic clock’s precision.

Laser technology has helped to better control the atoms, such as with optical lattices that can layer atoms into “pancakes” or egg-carton-like structures, immobilize them and helping eliminate Doppler shifts altogether. [Coming Soon: An Atomic Clock That Can Fit in Your Pocket ]

The official “rhythm” associated with the energy difference between the ground state and excited state of those cesium atoms, better known as the atomic transition frequency, yields something equivalent to the official definition of a second: 9,192,631,770 cycles of the radiation that gets a cesium atom to vibrate between those two energy states.

Future atomic clocks

Today’s atomic clocks mostly still use cesium, so according to NSF-funded physicist Kurt Gibble at Pennsylvania State University, the biggest advance in future atomic clocks will be a switch from measuring atoms vibrating at microwave frequencies to those vibrating at optical frequencies.

Today’s atomic clocks in GPS satellites, cell phone towers, the U.S. Naval Observatory’s master clock, and many other places in the world are microwave frequency clocks. These are the only clocks, at this point, that keep reliable time, Gibble said. Despite promising significantly more accuracy. “Just the higher frequency makes it a lot easier to be more accurate,” he added. “So far, optical standards don’t run for long enough to keep time, but they will soon.”

Gibble has an international reputation for assessing accuracy and improving microwave frequency clocks, including some of the most accurate clocks in the world: the cesium clocks at the United Kingdom’s National Physical Laboratory and the Observatory in Paris in France. He is now exploring new optical clocks that could further improve this field.

Optical frequency clocks actually operate on a significantly higher frequency than the microwave ones, which is why many researchers are exploring their potential with different atoms, including alkaline rare earth elements such as ytterbium, strontium and gadolinium.

Simien, whose research focuses on gadolinium, has studied minimizing or eliminating (if possible) key issues that limit accuracy. And recently, Gibble started work on another promising candidate, cadmium.

“Nowadays, the biggest obstacle, in my opinion, is the black body radiation shift,” Simien said. “The black body radiation shift is a systematic effect. We live in a thermal environment, meaning its temperature fluctuates. Even back in the day, a mechanical clock had pieces that would heat up and expand or cool down and contract. A clock’s accuracy varied with its environment. Today’s system is no longer mechanical and has better technology, but it is still susceptible to a thermal environment’s effects. Gadolinium is predicted to have a significantly reduced black body relationship compared to other elements implemented and being proposed as new frequency standards.”

According to Gibble, optical clocks are so accurate they would lose less than a second in the age of the universe, 13.8 billion years. And while Simien and Gibble agree that optical frequency atomic clock research represents the next generation of atomic clocks, taking accuracy to the next level, they recognize that most people don’t really care if the Big Bang happened 13 billion years ago or 13 billion years ago plus one second.

“It’s important to understand that one more digit of accuracy is not always just fine tuning something that is probably already good enough,” said John Gillaspy, an NSF program director who reviews funding for atomic clock research for the agency’s physics division. “Extremely high accuracy can sometimes mean a qualitative breakthrough which provides the first insight into an entirely new realm of understanding — a revolution in science.”

“Around the middle of the previous century, Willis Lamb measured a tiny frequency shift which led theorists to reformulate physics as we know it (not to mention earning him a Nobel Prize),” Gillaspy elaborated. “At a conference just this week, I heard a scientist discuss his idea to harness the GPS network’s precise timing to hunt for Dark Matter, one of the most outstanding problems in science today. Who knows when the next breakthrough will come, and whether it will be in the first digit or the 10th?

“Unfortunately, most people cannot appreciate why more accuracy matters, as evidenced in a recent blog post aimed at physicists in this field. The commenter wrote: ‘You’ve managed to find the single most depressing scientific endeavor of all time: Spend years of research trying to make an ultra-precise clock more precise. If they succeed, only electrons will notice’….These scientists know that they are, in fact, doing the sort of work that can change the world.”

“Interstellar” and beyond

Atomic clock researchers point to GPS as the most visible application of the basic science they study, but it’s only one way this foundational work holds promise.

Many physicists expect it to provide insight that not only illuminates understanding of fundamental physics and general relativity, but also advances quantum computing, sensor development and other sensitive instrumentation that requires clever design to resist the natural force of gravity, magnetic and electrical fields, temperature and motion.

Financial analysts, too, share concerns about the millions that could be lost in worldwide markets due to ill-synchronized clocks. In fact, on June 30, 2015, at 7:59:59 p.m. EDT, the world adds what is known as a “leap second” to keep solar time within 1 second of atomic time. However, because history has shown that most clocks won’t do it correctly, many major financial markets are planning to shut down for a period of time around this leap second, since it’s happening in the middle of a business day in many parts of the world — there’s concern that millions could be lost in worldwide markets due to ill-synchronized clocks.

“The reason you want better clocks isn’t to get accurate time over a long period down to the second. It’s the importance of being able to measure small time differences,” Gibble said. “The GPS looks at the difference in time for light propagating from several GPS satellites. The thing to remember is that the speed of light is one foot per nanosecond. If you want to know where you are, several GPS satellites send out a signal — a radio broadcast that tells where the satellites are and what time the radio signal left the satellite. Your GPS receiver gets the signals and looks at the time differences of the signals, when they arrive compared to when they said they left.”

Getting a GPS to guide us in deserts, tropical forests, oceans and other areas where roads aren’t around to help as markers along the way, one needs clocks with nanosecondprecision in GPS satellites to keep us from getting lost.

“If you want to know where you are to a couple of feet, you need to have timing to a nanosecond — a billionth of a second, which is 10 to the minus 9 of a second,” added Gibble. “If you want that clock to be good for more than a day, then already you have to be at 10 to the minus 14. If you want the system to go for two weeks or longer, then you need something significantly better than that.”

And then there’s the future to think about.

“Remember the movie, ‘Interstellar’?” Simien asks. “There is someone on a spaceship far away, Matthew McConaughey is on a planet in a strong gravitational field. He experiences reality in terms of hours, but the other individual back on the spacecraft experiences years. That’s general relativity. Atomic clocks can test this kind of fundamental theory and its various applications that make for fascinating science, and as you can see, also expand our lives.”

References:http://www.livescience.com/

A human skull, on average, is about 6.8 millimeters (0.3 inches) thick, or roughly the depth of the latest smartphone. Human skin, on the other hand, is about 2 to 3 millimeters (0.1 inches) deep, or about three grains of salt deep. While both of these dimensions are extremely thin, they present major hurdles for any kind of imaging with laser light.

Why? The photons in laser light scatter when they encounter biological tissue. Corralling tiny photons to obtain meaningful details about tissue has proven to be one of the most challenging problems laser researchers have faced to date. However, researchers at Washington University in St. Louis (WUSTL) decided to eliminate the photon roundup completely and use scattering to their advantage. The result: an imaging technique that would peer right into a skull, penetrating tissue at depths up to 7 centimeters (about 2.8 inches).

However, researchers at Washington University in St. Louis (WUSTL) decided to eliminate the photon roundup completely and use scattering to their advantage. The result: an imaging technique that would peer right into a skull, penetrating tissue at depths up to 7 centimeters (about 2.8 inches).

The photoacoustic effect

The approach, which combines laser light and ultrasound, is based on the photoacoustic effect, a concept first discovered by Alexander Graham Bell in the 1880s. In his work, Bell discovered that the rapid interruption of a focused light beam produces sound.

To produce the photoacoustic effect, Bell focused a beam of light on a selenium block. He then rapidly interrupted the beam with a rotating slotted disk. He discovered that this activity produced sound waves. Bell showed that the photoacoustic effect depended on the absorption of light by the block, and the strength of the acoustic signal depended on how much light the material absorbed.

“We combine some very old physics with a modern imaging concept,” said WUSTL researcher Lihong Wang, who pioneered the approach. Wang and his WUSTL colleagues were the first to describe functional photoacoustic tomography (PAT) and 3D photoacoustic microscopy (PAM). [Listening with Lasers: Hybrid Technique Sees Into Human Body ]

The two techniques follow the same basic principles: When the researchers shine a pulsed laser beam into biological tissue, the beam spreads out and generates a small, but rapid rise in temperature. This produces sound waves that are detected by conventional ultrasound transducers. Image reconstruction software converts the sound waves into high-resolution images.

Following a tortuous path

Wang began exploring the combination of sound and light as a postdoctoral researcher. At the time, he developed computer models of photons as they traveled through biological material. This work led to an NSF Faculty Early Career Development (CAREER) grant to study ultrasound encoding of laser light to “trick” information out of the laser beam.

Unlike other optical imaging techniques, photoacoustic imaging detects ultrasonic waves induced by absorbed photons, no matter how many times the photons have scattered. Multiple external detectors capture the sound waves regardless of their original locations. “While the light travels on a highly tortuous path, the ultrasonic wave propagates in a clean and well-defined fashion,” said Wang. “We see optical absorption contrast by listening to the object.”

Because the approach does not require injecting imaging agents, researchers can study biological material in its natural environment. Using photoacoustic imaging, researchers can visualize a range of biological material, from cells and their component parts to tissue and organs. Scientists can even detect single red blood cells in blood, or fat and protein deposits in arteries.

While PAT and PAM are primarily used in laboratory settings, Wang and others are working on multiple clinical applications. In one example, researchers use PAM to study the trajectory of blood cells as they flow through vessels in the brain.

“By seeing individual blood cells, researchers can start to identify what’s happening to the cells as they move through the vessels. Watching how these cells move could act as an early warning system to allow detection of potential blockage sites,” said Richard Conroy, director of the Division of Applied Science and Technology at the U.S. National Institute of Biomedical Imaging and Bioengineering.

Minding the gap

Because PAT and PAM images can be correlated with those generated using other techniques, such as magnetic resonance imaging (MRI) or positron emission tomography (PET), these techniques are complementary. “One imaging modality can’t do everything,” said Conroy. “Comparing results from different modalities provides a more detailed understanding of what is happening from the cell level to the whole animal.”

The approach could help bridge the gap between animal and human research, especially in neuroscience.

“Photoacoustic imaging is helping us understand how the mouse brain works,” said Wang. “We can then apply this information to better understand how the human brain works.” Wang, along with his team, is applying both PAT and PAM to study mouse brain function.

One of the challenges currently facing neuroscientists is the lack of available tools to study brain activity, Wang said. “The holy grail of brain research is to image action potentials,” said Wang. (An action potential occurs when electrical signals travel along axons, the long fibers that carry signals away from the nerve cell body.) With funding from the U.S. BRAIN Initiative , Wang and his group are now developing a PAT system to capture images every one-thousandth of a second, fast enough to image action potentials in the brain.

“Photoacoustic imaging fills a gap between light microscopy and ultrasound,” said Conroy. “The game-changing aspect of this [Wang’s] approach is that it has redefined our understanding of how deep we can see with light-based imaging,” said Conroy.

References:http://www.livescience.com/

Timelapse image of Shakey in action

Robots are increasingly becoming part of our everyday lives and many roboticists believe that we are on the verge of a robot revolution that will do for goods and services what the Internet did for information. If so, then a lot of the credit goes to a 50 year old box on wheels called Shakey: the “world’s first electronic person.”

In the mid-1960s, computers were undergoing the first big jump in their evolution since “electronic brains” became practical in the 1940s. Computer architecture was much more sophisticated, scientists and engineers had a better grasp of the technology, transistors were shrinking mainframes so they filled a room instead of a really big room, and some researchers were convinced that computers and robots would soon be able to enter real world settings.

Among these was Charlie Rosen, one of the pioneers in the field of artificial intelligence and founder of SRI International’s Artificial Intelligence Center (then the Stanford Research Institute). He believed that computer simulations had reached the stage where they made it possible to produce problem-solving robots capable of working in factories.

 

At the same time ARPA, the ancestor of today’s DARPA, was interested in finding ways to use robots and artificial intelligence for military reconnaissance. One idea was that it might be possible to build some sort of a robotic scout for the Army, so, along with the National Science Foundation, and the Office of Naval Research, the agency hired SRI to look into such a scout.

The result was Shakey.

Officially, the timeline for project Shakey ran from 1966 to 1972, but like all complicated endeavors, the origins go back a bit further as the SRI team, consisting of project manager Rosen, Peter Hart, Marty Tenenbaum, Nils Nilsson, and others, gathered and drew up outlines and proposals for the project. Hart recently gave a keynote address at the ICRA conference in Seattle calling this year the 50th anniversary – and he was there.

Dates aside, what was remarkable about the Shakey project wasn’t just what it accomplished, but also but what its ambitions showed about the state of robotics. What Rosen et. al. set out to do indicated that they were either decades ahead of their time, or were in way over their heads.

The purpose of Shakey wasn’t just to produce a robot, but also one that was mobile, autonomous, equipped with vision, capable of mapping its surroundings, could solve problems, and could be addressed in natural-language English. Even today, combining only a fraction of these would ambitious. To combine them all using 1966 technology would have given Dr Frankenstein pause, but Shakey went on to become the first robot to combine motion, perception, and problem solving in one mobile package.

But why was it called Shakey? The answer was in its construction. Looking like a very early draft of R2D2, Shakey was a stack of gear. On the bottom was a motorized platform containing the drive wheels, push bar, and cat’s whiskers bump sensors. Above this was the electronic components box, then the “head” consisting of a black and white television camera, microphone, and a spinning prism rangefinder with the radio antenna on the very top. This arrangement was practical, but not very stable, hence the name.

According to Rosen: “We worked for a month trying to find a good name for it, ranging from Greek names to whatnot, and then one of us said, ‘Hey, it shakes like hell and moves around, let’s just call it Shakey,.'”

Even standing about five feet tall, Shakey didn’t seem very large until someone informed you that he (the team always called it “he”) actually weighed several tons. The visible part was just his mobile extension. The actual brain filled an entire room in another part of the lab with a second computer acting as a control interface. Originally, the mainframe was a 64K SDS-940 computer programmed in Fortran and Lisp, which was later replaced with a 192K PDP-10 around 1969. To put that into perspective, there are handheld games today with much more power than both of these put together.

Shakey lived in his own little world, which was made up of a series of rooms, doors, and objects. The walls had baseboards and everything was painted white and red to provide contrast for the monochromatic vision system. The robot’s job was to navigate these rooms, planning how to carrying out tasks, and solving any problems that might arise,

Early versions of Shakey envisioned manipulators, but this was later deemed unnecessarily complicated, so the final design worked by pushing objects around.

Much like other computers of the day, Shakey was given commands by teletype and responded via teletype and cathode ray tube. The robot was controlled by a tiered series of actions, which allowed Shakey to assess situations and find answers to problems.

At the bottom were low-level action programs, which were given to Shakey in plain English with commands like “go,” “pan,” and “tilt.” These were automatically translated into predicate calculus and provided the building blocks for intermediate-level actions, such as “go to” that told the robot to go to a particular location.

Telling Shakey where to go may seem simple, but if the location was in another room or behind an obstacle, that required him to set up subgoals or way points with a route to reach them. It was even more complicated if Shakey was told to find an object and move it somewhere else.

Shakey navigated by dead reckoning. That is, he counted turns of the drive wheels and deduced his location. However, this accumulated errors, so Shakey had to back it up with visual information. His pattern recognition algorithms could pick things like object outlines or room corners, allowing him to build up a map of his world and adjust his navigation.

The key to this was STRIPS (Stanford Research Institute Problem Solver), which is an artificial intelligence language for automated planning that allowed Shakey to map actions and subgoals to chart a plan of action. This also gave him the ability to recover from accidents, such as a misplaced box, an unknown room, an unexpected obstacle, or an obstructed door, and planning moves to recover from it, as well as learning from past mistakes by combining commands, intermediate actions, and preconditions to produce new actions.

Eventually, Shakey was able to solve surprisingly complex puzzles, such as the monkey and banana problem. In this, Shakey had to move a box on to a platform, which meant he had to figure out how to move a ramp into position first. It took the robot days to accomplish with many false starts and repetitions, but at the time, it was an unprecedented example of robotics and artificial intelligence.

The reaction to Shakey by the public and even by scientists was one of sensation. Shakey was described in Life magazine in 1970 as “the first electronic person,” and thinkers like Marvin Minsky were seriously predicting true artificial intelligence of superhuman power within three to eight years on the strength of Shakey’s performance.

If nothing else, it’s a cautionary tale about overestimating the state of the art. Bear in mind that this was a time when it was commonly believed that an invincible chess-playing computer would be built any day now instead of decades into the future – contemporary chess computers were lucky if they could manage a legal game, much less win one.

The Shakey project carried on until 1972 when the Defense Department started to get impatient for results. One general was even reported to ask: “Can you mount a 36-inch bayonet on it?” Unfortunately, despite remarkable advances, Shakey was still a creature of the laboratory and funding dried up.

Today, Shakey is on display at the Computer History Museum in Mountain View, California. However, it’s fair to say that without Shakey, today’s robots wouldn’t have been possible.
“Shakey, the first mobile robot to reason about its actions, was groundbreaking not only to robotics, but to artificial intelligence as well, as it led to fundamental advances in visual analysis, route finding, and planning of complex actions,” says Ray Perrault, PhD, director of SRI International’s Artificial Intelligence Center. “Shakey also helped open the possibilities of computer science to the public’s imagination, and put SRI’s Artificial Intelligence Center on the map.”

More information about Shakey is available at SRI International and a video showing the robot in action can be found here.

References:http://www.gizmag.com/