Intel

Intel's Pohoiki Beach is a Neuromorphic Computer Capable of Simulating 8 Million Neurons (venturebeat.com) 58

During the Defense Advanced Research Projects Agency's (DARPA) Electronics Resurgence Initiative 2019 summit in Detroit, Michigan, Intel unveiled a system codenamed "Pohoiki Beach," a 64-chip computer capable of simulating 8 million neurons in total. From a report: Intel Labs managing director Rich Uhlig said Pohoiki Beach will be made available to 60 research partners to "advance the field" and scale up AI algorithms like spare coding and path planning. [...] Pohoiki Beach packs 64 128-core, 14-nanometer Loihi neuromorphic chips, which were first detailed in October 2017 at the 2018 Neuro Inspired Computational Elements (NICE) workshop in Oregon.

They have a 60-millimeter die size and contain over 2 billion transistors, 130,000 artificial neurons, and 130 million synapses, in addition to three managing Lakemont cores for task orchestration. Uniquely, Loihi features a programmable microcode learning engine for on-chip training of asynchronous spiking neural networks (SNNs) -- AI models that incorporate time into their operating model, such that components of the model don't process input data simultaneously. This will be used for the implementation of adaptive self-modifying, event-driven, and fine-grained parallel computations with high efficiency.

Science

A Cell Tower In the Swiss Alps Is Struck By Lightning More Than 100 Times a Year (ieee.org) 90

Wave723 quotes IEEE Spectrum: Atop a rocky peak in the Swiss Alps sits a telecommunications tower that gets struck by lightning more than 100 times a year, making it perhaps the world's most frequently struck object. Taking note of the remarkable consistency with which lightning hits this 124-meter structure, researchers have adorned it with instruments for a front-row view of these violent electric discharges...

To anyone who has witnessed a lightning strike, everything seems to happen all at once. But [New Mexico Tech's Mark] Stanley's sensor captures several gigabytes of data about the many separate pulses that occur within each flash. Those data can be made into a video that replays, microsecond by microsecond, how "channels" of lightning form in the clouds.... [T]hey intend to use data gathered by the tower's many instruments (which include a collection of six antennas called a lightning mapping array, two Rogowski coils to measure current, two B-Dot sensors to measure the current time-derivative, broadband electric and magnetic field sensors, and a high-speed camera) to reconstruct the total path of strikes soon after they happen, tracing the electromagnetic radiation all the way back to its original source...

The Santis team's work has held particular relevance for wind farm operators. That's because most strikes recorded at the tower are examples of upward lightning -- which travels from ground-to-cloud instead of cloud-to-ground.

They hope to eventually help make progress on predicting where lightning will strike.

And by the end of this year, the team at the tower expect to record their 1,000th lightning strike.
NASA

NASA Funds Company To 3D-Print Spacecraft Parts in Orbit (engadget.com) 28

An anonymous reader quotes Engadget: NASA is expanding its efforts to bring 3D printing to space. The agency has given Made In Space a $73.3 million contract to demonstrate the ability to 3D-print spacecraft parts in orbit using Archinaut One, a robotic manufacturing ship due to launch in 2022 or later. The vessel will fly aboard a Rocket Lab Electron rocket and 3D-print two 32-foot beams on each side, with each unfurling two solar arrays. The completed arrays could produce up to five times more power than the solar panels you normally find on spacecraft this size, NASA said...

If successful, it could alter how NASA and others approach building and fixing spacecraft. This could lead to building spacecraft (albeit smaller ones at this stage) in orbit, of course, but it could also let space agencies launch small satellites that receive large power collectors once they're floating above Earth. It could also lead to fewer spacewalks by having robots build items that would otherwise require human involvement.

Moon

India Plans Historic Launch of a Rocket to the Moon (techcrunch.com) 89

The Indian Space Research Organization (ISRO) is planning to launch a rocket to the moon. UPDATE: 2:28: The launch "has been called off for today," the ISRO posted on Twitter.

But when the re-scheduled launch happens, you can watch it live on two YouTube channels, on Twitter, Facebook, or webcast on the ISRO's web site.

TechCrunch has also embedded a livestreaming video in their report: Chandrayaan-2 will carry lunar lander Vikram, which will deliver ISRO rover Pragyan to the surface at the pole, with a target landing zone of a plain that covers the ground between two of the Moon's craters, Simpelius N and Manzinus C. The rocket used for the launch is the GSLV Mk-III, India's most powerful launch vehicle ever, and the orbiter used for this mission will relay back information from the lander and rover to Earth via the Indian Deep Space Network, as well as make its own observations during its planned one-year mission lifespan.

The mission will seek to take a number of measurements of the lunar surface, including topographic, mineral makeup, seismographic, chemical analytics and more, with an eye to shedding more light on the Moon's origins. If all goes to plan, the lunar orbiter will make its way to to Moon over the next couple of months and aim to soft land the Vikram at the South Pole target site on September 6, 2019.

This is a historic mission for a few reasons, including being the first ever soft-landing attempt at the Moon's South Pole region, as well as being the first Indian mission to attempt a soft landing using all home-grown lander and rover technology. If successful, India will be only the fourth country ever to have soft-landed a vehicle on the lunar surface.

Medicine

Doing Five Things Could Decrease Your Risk of Alzheimer's By 60% (sfgate.com) 132

"Light-to-moderate" alcohol consumption can help reduce your risk of Alzheimer's disease.

An anonymous reader quotes the Washington Post: A study presented Sunday at the Alzheimer's Association International Conference in Los Angeles found that combining five lifestyle habits -- including eating healthier, exercising regularly and refraining from smoking -- can reduce the risk of Alzheimer's by 60 percent. A separate study showed that lifestyle choices can lower risk even for those who are genetically prelifestyle disposed to the disease...

Over the last decade, studies have increasingly pointed to controllable lifestyle factors as critical compenents to reducing the risk of cognitive decline. Researchers say that, as with heart disease, combating dementia will probably require a "cocktail" approach combining drugs and lifestyle changes. And as recent efforts to develop a cure or more effective drug treatments for dementia have proven disappointing, the fact that people can exert some control in preventing the disease through their own choices is encouraging news, they say.

While the new study's authors expected to see that leading a healther life decreases the chance of dementia, they were floored by the "magnitude of the effect," said Klodian Dhana, a Rush University professor and co-author. "This demonstrates the potential of lifestyle behaviors to reduce risk as we age," said Heather Snyder, senior director of medical and scientific operations at the Alzheimer's Association. "The fact that four or five lifestyle habits put together can have that kind of benefit for your brain is incredibly powerful."

The fifth lifestyle habit is "engaging in mentally stimulating activities like reading the newspaper, visiting the library or playing games such as chess and checkers."

Time reports that even following just two or three of the healthy lifestyle factors reduced the risk of developing Alzheimer's dementia in the study by at least 39%.
Science

IQ Test Scores Increased For a Century. But Did it Help? (bbc.com) 260

IQ test scores have been increasing for 100 years, reports a senior journalist at BBC Future. He also writes that there's evidence "that we may have already reached the end of this era -- with the rise in IQs stalling and even reversing."

But this raises an even larger question: did a century of increasing scores on IQ tests bring benefits to society? You might assume that the more intelligent you are, the more rational you are, but it's not quite this simple... Consider the abundant literature on our cognitive biases. Something that is presented as "95% fat-free" sounds healthier than "5% fat", for instance -- a phenomenon known as the framing bias. It is now clear that a high IQ does little to help you avoid this kind of flaw, meaning that even the smartest people can be swayed by misleading messages. People with high IQs are also just as susceptible to the confirmation bias -- our tendency to only consider the information that supports our pre-existing opinions, while ignoring facts that might contradict our views. That's a serious issue when we start talking about things like politics.

Nor can a high IQ protect you from the sunk cost bias -- the tendency to throw more resources into a failing project, even if it would be better to cut your losses -- a serious issue in any business. (This was, famously, the bias that led the British and French governments to continue funding Concorde planes, despite increasing evidence that it would be a commercial disaster.) Highly intelligent people are also not much better at tests of "temporal discounting", which require you to forgo short-term gains for greater long-term benefits. That's essential, if you want to ensure your comfort for the future.

Besides a resistance to these kinds of biases, there are also more general critical thinking skills -- such as the capacity to challenge your assumptions, identify missing information, and look for alternative explanations for events before drawing conclusions. These are crucial to good thinking, but they do not correlate very strongly with IQ, and do not necessarily come with higher education. One study in the USA found almost no improvement in critical thinking throughout many people's degrees. Given these looser correlations, it would make sense that the rise in IQs has not been accompanied by a similarly miraculous improvement in all kinds of decision making.

The article concludes that "this kind of thinking can be taught -- but it needs deliberate and careful instruction," and suggests "we might also make a more concerted and deliberate effort to improve those other essential skills too that do not necessarily come with a higher IQ..."

"Ideally, we might then start to see a steep rise in rationality -- and even wisdom... If so, the temporary blip in our IQ scores need not represent the end of an intellectual golden age -- but its beginning."
AI

Amazon's Alexa Will Deliver NHS Medical Advice in the UK (theverge.com) 45

"The UK's National Health Service (NHS) has announced what it claims is a world first: a partnership with Amazon's Alexa to offer health advice from the NHS website."

An anonymous reader quotes the Verge: Britons who ask Alexa basic health questions like "Alexa, how do I treat a migraine?" and "Alexa, what are the symptoms of flu?" will be given answers vetted by NHS health professionals and currently available on its website. At the moment, Alexa sources answers to such questions from a variety of places, including the Mayo Clinic and WebMD.

The partnership does not add significantly to Alexa's skill-set, but it is an interesting step for the NHS. The UK's Department of Health (DoH) says it hopes the move will reduce the pressure on health professionals in the country, giving people a new way to access reliable medical advice. It will also benefit individuals with disabilities, like sight impairments, who may find it difficult to use computers or smartphones to find the same information.

Space

Inside 'Starshot', the Audacious Plan To Shoot Tiny Ships To Alpha Centauri (technologyreview.com) 229

"Starshot wants to build the world's most powerful laser and aim it at the closest star. What could go wrong?"

An anonymous reader quotes MIT's Technology Review: In 2015, Philip Lubin, a cosmologist from the University of California, Santa Barbara, took the stage at the 100-Year Starship Symposium in Santa Clara. He outlined his plan to build a laser so powerful that it could accelerate tiny spacecraft to 20% of the speed of light, getting them to Alpha Centauri in just 20 years. We could become interstellar explorers within a single generation. It was quite the hook.

Because Lubin is an excellent public speaker, and because the underlying technologies already existed, and because the science was sound, he was mobbed after the talk. He also met Pete Worden, a former research director of NASA's Ames Research Center, for the first time. Worden had recently taken over as head of the Breakthrough Initiatives, a nonprofit program funded by Russian technology billionaire Yuri Milner. Six months later, Lubin's project had $100 million in funding from Breakthrough and the endorsement of Stephen Hawking, who called it the "next great leap into the cosmos."

Starshot is straightforward, at least in theory. First, build an enormous array of moderately powerful lasers. Yoke them together—what's called "phase lock"—to create a single beam with up to 100 gigawatts of power. Direct the beam onto highly reflective light sails attached to spacecraft weighing less than a gram and already in orbit. Turn the beam on for a few minutes, and the photon pressure blasts the spacecraft to relativistic speeds.

Not only could such a technology be used to send sensors to another star system; it could dispatch larger craft to Earth's neighboring planets and moons. Imagine a package to Mars in a few days, or a crewed mission to Mars in a month. Starshot effectively shrinks the solar system, and ultimately the galaxy.

It's fantastic. And also a dream. Or a sales pitch. Or a long-term, far-out project that can't be sustained long enough for the nonexistent technologies it requires to be built.

Science

Carbon Nanotube Device Channels Heat Into Light (phys.org) 102

An anonymous reader quotes a report from Phys.Org: Rice University scientists are designing arrays of aligned single-wall carbon nanotubes to channel mid-infrared radiation (aka heat) and greatly raise the efficiency of solar energy systems. Their invention is a hyperbolic thermal emitter that can absorb intense heat that would otherwise be spewed into the atmosphere, squeeze it into a narrow bandwidth and emit it as light that can be turned into electricity. The aligned nanotube films are conduits that absorb waste heat and turn it into narrow-bandwidth photons. Because electrons in nanotubes can only travel in one direction, the aligned films are metallic in that direction while insulating in the perpendicular direction, an effect called hyperbolic dispersion. Thermal photons can strike the film from any direction, but can only leave via one. Adding the emitters to standard solar cells could boost their efficiency from the current peak of about 22%. "By squeezing all the wasted thermal energy into a small spectral region, we can turn it into electricity very efficiently," he said. "The theoretical prediction is that we can get 80% efficiency." The study has been published in the journal ACS Photonics.
Science

A New Study Uses Camera Footage To Track the Frequency of Bystander Intervention (citylab.com) 67

An anonymous reader quotes a report from CityLab: It's one of the most enduring urban myths of all: If you get in trouble, don't count on anyone nearby to help. Research dating back to the late 1960s documents how the great majority of people who witness crimes or violent behavior refuse to intervene. Psychologists dubbed this non-response as the "bystander effect" -- a phenomenon which has been replicated in scores of subsequent psychological studies. The "bystander effect" holds that the reason people don't intervene is because we look to one another. The presence of many bystanders diffuses our own sense of personal responsibility, leading people to essentially do nothing and wait for someone else to jump in.

Past studies have used police reports to estimate the effect, but results ranged from 11 percent to 74 percent of incidents being interventions. Now, widespread surveillance cameras allow for a new method to assess real-life human interactions. A new study published this year in the American Psychologist finds that this well-established bystander effect may largely be a myth. The study uses footage of more than 200 incidents from surveillance cameras in Amsterdam; Cape Town; and Lancaster, England. The study finds that in nine out of 10 incidents, at least one bystander intervened, with an average of 3.8 interveners. There was also no significant difference across the three countries and cities, even though they differ greatly in levels of crime and violence.
The study actually found that the more bystanders there were, the more likely it was that at least someone would intervene to help. "This is a powerful corrective to the common perception of 'stranger danger' and the 'unknown other,'" reports CityLab. "It suggests that people are willing to self-police to protect their communities and others."
Medicine

Billions of Air Pollution Particles Found in Hearts of City Dwellers (theguardian.com) 56

The hearts of young city dwellers contain billions of toxic air pollution particles, research has revealed. The Guardian: Even in the study's youngest subject, who was three, damage could be seen in the cells of the organ's critical pumping muscles that contained the tiny particles. The study suggests these iron-rich particles, produced by vehicles and industry, could be the underlying cause of the long-established statistical link between dirty air and heart disease. The scientists said the abundance of the nanoparticles might represent a serious public health concern and that particle air pollution must be reduced urgently. More than 90% of the world's population lives with toxic air, according to the World Health Organization, which has declared the issue a global "public health emergency."

The scientists acknowledged some uncertainties in their research, but Prof Barbara Maher, of Lancaster University, said: "This is a preliminary study in a way, but the findings and implications were too important not to get the information out there." Maher and colleagues found in 2016 that the same nanoparticles were present in human brains and were associated with Alzheimers-like damage, another disease linked to air pollution. While all ages were affected, Maher said she was particularly concerned about children.

Medicine

Fatal Accident With Metal Straw Highlights a Risk (nytimes.com) 328

The disturbing death of a woman in Britain renewed a debate that has followed bans on plastic straws around the world. From a report: A British woman was impaled by a metal straw after falling at her home, a coroner said in an inquest this week that highlighted the potential dangers of metal straws. Such straws have surged in popularity as cities, states and even countries have banned single-use plastic straws. A British straw ban will go into effect in April, but the worldwide environmental push against single-use straws has encountered opposition from some caregivers and advocates for people with disabilities. They have voiced worries about the safety of rigid straws and the overall availability of straws for people who are unable to drink without them. [...] Many people with disabilities rely on straws to drink, Ms. Sauder said, but could have difficulties finding them in states and cities, such as California and Seattle, that have banned or restricted single-use straws.

Starbucks plans to eliminate its ubiquitous green plastic straws at 28,000 of its locations around the world in 2020. It's not easy being green for Starbucks, however. In 2016, the coffee chain recalled stainless steel straws sold at its shops because they posed an injury risk. At the time, Starbucks said it had received reports of three children in the United States and one in Canada who had been lacerated by the straws, which were sold with reusable beverage containers. Dentists say that improper use of metal or glass straws can also be bad for teeth. "Clearly, chewing on a metal or glass straw can be hazardous to your teeth and your health," said Dr. Timothy Chase of SmilesNY Cosmetic and Implant Dentistry in New York. "Just like we tell people not to chew on pens." Dr. Chase added that it's important to keep reusable straws clean to avoid infection-causing bacteria.

Earth

Streaming Online Pornography Produces as Much CO2 as Belgium (newscientist.com) 169

The transmission and viewing of online videos generates 300 million tonnes of carbon dioxide a year, or nearly 1 per cent of global emissions. On-demand video services such as Netflix account for a third of this, with online pornographic videos generating another third. From a report: This means the watching of pornographic videos generates as much CO2 per year as is emitted by countries such as Belgium, Bangladesh and Nigeria. That's the conclusion of a French think tank called The Shift Project. Earlier this year, it estimated that digital technologies produce 4 per cent of greenhouse gas emissions and that this figure could soar to 8 per cent by 2025. Now it has estimated the CO2 emissions due to online videos alone. The report's authors used 2018 reports by companies Cisco and Sandvine to work out global video internet traffic. They then estimated how much electricity was used to carry this video data and view it on different devices, from phones to TVs. Finally, they estimated the overall emissions using global average figures for carbon emissions from electricity generation. Online video accounted for 60 per cent of global data flows in 2018, the report states, or 1 zettabyte of data (one thousand billion billion bytes).
Media

Scientists Use Camera With Human-Like Vision To Capture 5,400 FPS Video (petapixel.com) 66

An anonymous reader quotes a report from PetaPixel: A team of scientists from the Swiss Federal Institute of Technology in Zurich (ETH Zurich) have figured out how to capture super slow-motion footage using what's called an "Event Camera." That is: a camera that sees the world in a continuous stream of information, the way humans do. Regular cameras work by capturing discrete frames, recapturing the same scene 24 or more times per second and then stitching it together to create a video. Event cameras are different. They capture "pixel-level brightness changes" as they happen, basically recording each individual light "event" as it happens, without wasting time capturing all the stuff that remains the same frame by frame.

As ETH Zurich explains, some of the advantages of this type of image capture is "a very high dynamic range, no motion blur, and a latency in the order of microseconds." The downside is that there's no easy way to process the resulting "footage" into something you can display using current algorithms because they all expect to receive a set of discrete frames. Well, there was no easy way. This is what the folks at ETH Zurich just improved upon, developing a reconstruction model that can interpret the footage to the tune of 5,000+ frames per second. The results are astounding: a 20% increase in the reconstructed image quality over any model that existed before, and the ability to output "high frame rate videos (more than 5,000 frames per second) of high-speed phenomena (e.g. a bullet hitting an object)," even in high dynamic range "challenging lighting conditions."
Their findings have been published in a research paper titled High Speed and High Dynamic Range Video with an Event Camera.
Facebook

Facebook AI Pluribus Defeats Top Poker Pros In 6-Player Texas Hold 'Em (arstechnica.com) 58

Carnegie Mellon University and Facebook AI research scientists have developed an AI dubbed Pluribus that took on 15 professional human players in six-player no-limit Texas Hold 'em and won. The researchers describe how they achieved this feat in a new paper in Science. Ars Technica reports: Playing more than 5,000 hands each time, five copies of the AI took on two top professional players: Chris "Jesus" Ferguson, six-time winner of World Series of Poker events, and Darren Elias, who currently holds the record for most World Poker Tour titles. Pluribus defeated them both. It did the same in a second experiment, in which Pluribus played five pros at a time, from a pool of 13 human players, for 10,000 hands.

Co-author Tuomas Sandholm of Carnegie Mellon University has been grappling with the unique challenges poker poses for AI for the last 16 years. No-Limit Texas Hold 'em is a so-called "imperfect information" game, since there are hidden cards (held by one's opponents in the hand) and no restrictions on the size of the bet one can make. By contrast, with chess and Go, the status of the playing board and all the pieces are known by all the players. Poker players can (and do) bluff on occasion, so it's also a game of misleading information.

Slashdot Top Deals