What Happened to the Holy Grail?

Underground Stream - 12 hours 46 min ago

The myth of the Holy Grail is embedded in our culture - the sacred Christian treasure, and the mystery of where it disappeared to (or even if it existed in the first place, as many historians believe it is simply a legend), has fascinated us for so long that it has now even become a turn of phrase signifying the greatest search. It's found in everything from the Grail romances to Dan Brown (even if, in the latter case, as 'the royal blood', rather than an actual cup).

But is it possible that the legendary religious treasure does exist? The short video below, posted by the Smithsonian Museum, discusses the research of Spanish historian Marguerita Torres, who may have found out where the cup went after it disappeared from the Church of the Holy Sepulchre in 909 AD. In a medieval text found in a library in Egypt, Torres discovered that 150 years after going missing - possibly looted during 11th century troubles in the Holy Land - the Grail may have made its way into the hands of a Spanish king:

The text tells the story of a Muslim entourage giving the cup to a king in Spain - the King of Leon, Ferdinand the Great. But if the cup was looted, why would it then be handed to a Christian king in Spain? In 1055, Spain was divided. Muslims rule the south, and Christians the north. Leon is the most powerful Christian kingdom. And its king, Ferdinand, is looking to push south, into Muslim kingdoms. The document reveals that a Muslim ruler...gave the cup of Christ to King Ferdinand in a bid to prevent any possible invasion.

From this evidence, Marguerita Torres believes she has identified the Holy Grail: the goblet of the Infanta Doña Urraca, found in the museum of the Basilica of San Isidoro in Leon, Spain. Historians disagree. But if you're interested in hearing more about Torres' argument, check out her book Kings of the Grail: Tracing the Historic Journey of the Cup of Christ from Jerusalem to Modern-Day Spain (Amazon US/Amazon UK).
.

A machine-learning system that trains itself by surfing the web

Kurzweil AI - 13 hours 6 min ago

MIT researchers have designed a new machine-learning system that can learn by itself to extract text information for statistical analysis when available data is scarce.

This new “information extraction” system turns machine learning on its head. It works like humans do. When we run out of data in a study (say, differentiating between fake and real news), we simply search the Internet for more data, and then we piece the new data together to make sense out of it all.

That differs from most machine-learning systems, which are fed as many training examples as possible to increase the chances that the system will be able to handle difficult problems by looking for patterns compared to training data.

Andrew Ng, Associate Professor of Computer Science at Stanford, Chief Scientist of Baidu, and Chairman and Co-founder of Coursera, is writing an introductory book, Machine Learning Yearning, intended to help readers build highly effective AI and machine learning systems. If you want to download a free draft copy of each chapter as it is finished (and previous chapters), you can sign up here for his mailing list. Ng is currently up to chapter 14.

“In information extraction, traditionally, in natural-language processing, you are given an article and you need to do whatever it takes to extract correctly from this article,” says Regina Barzilay, the Delta Electronics Professor of Electrical Engineering and Computer Science and senior author of a new paper presented at the recent Association for Computational Linguistics’ Conference on Empirical Methods on Natural Language Processing.

“That’s very different from what you or I would do. When you’re reading an article that you can’t understand, you’re going to go on the web and find one that you can understand.”

Confidence boost from web data

Machine learning systems determine whether they have enough data by assigning each of its classifications a confidence score — a measure of the statistical likelihood that the classification is correct, given the patterns discerned in the training data. If not, additional training data is required.

Fig 1. Sample news article on a shooting case. Note how the article contains both the name of the shooter and the number of people killed, but both pieces of information require complex extraction schemes to make sense out of the information. (credit: Karthik Narasimhan et al.)

In the real world, that’s not always easy. For example, the researchers note in the paper that the example news article excerpt in Fig. 1, “does not explicitly mention the shooter (Scott Westerhuis), but instead refers to him as a suicide victim. Extraction of the number of fatally shot victims is similarly difficult as the system needs to infer that ‘A couple and four children’ means six people. Even a large annotated training set may not provide sufficient coverage to capture such challenging cases.”

Instead, with the researchers’ new information-extraction system, if the confidence score is too low, the system automatically generates a web search query designed to pull up texts likely to contain the data it’s trying to extract. It then attempts to extract the relevant data from one of the new texts and reconciles the results with those of its initial extraction. If the confidence score remains too low, it moves on to the next text pulled up by the search string, and so on.

The system learns how to generate search queries, gauge the likelihood that a new text is relevant to its extraction task, and determine the best strategy for fusing the results of multiple attempts at extraction.

Testing the new information-extraction system

The MIT researchers say they tested their system with two information-extraction tasks. In each case, the system was trained on about 300 documents. One task was focused on collecting and analyzing data on mass shootings in the U.S. (an essential resource for any epidemiological study of the effects of gun-control measures).

The other task was the collection of similar data on instances of food contamination. It extracted food type, type of contaminant, and location.

For the mass-shootings task, the researchers asked the system to extract from websites (such as news articles, as in Fig. 1) the name of the shooter, the location of the shooting, the number of people wounded, and the number of people killed.

The goal was to find other documents that contain the information sought, expressed in a form that a basic extractor can “understand.”

Fig. 2. Two other articles on the same shooting case. The first article clearly mentions that six people were killed. The second one portrays the shooter in an easily extractable form. (credit: Karthik Narasimhan et al.)

For instance, Figure 2 shows two other articles describing the same event in which the entities of interest — the number of people killed and the name of the shooter — are expressed explicitly. That simplifies things.

From those articles, the system learned clusters of search terms that tended to be associated with the data items it was trying to extract. For instance, the names of mass shooters were correlated with terms like “police,” “identified,” “arrested,” and “charged.” During training, for each article the system was asked to analyze, it pulled up, on average, another nine or 10 news articles from the web.

The researchers then compared their system’s performance to that of several extractors trained using more conventional machine-learning techniques. For every data item extracted in both tasks, the new system outperformed its predecessors, usually by about 10 percent.

“The challenges … lie in (1) performing event coreference (retrieving suitable articles describing the same incident) and (2) reconciling the entities extracted from these different documents,” the authors note in the paper. “We address these challenges using a Reinforcement Learning (RL) approach that combines query formulation, extraction from new sources, and value reconciliation.”

Abstract of Improving Information Extraction by Acquiring External Evidence with Reinforcement Learning

Most successful information extraction systems operate with access to a large collection of documents. In this work, we explore the task of acquiring and incorporating external evidence to improve extraction accuracy in domains where the amount of training data is scarce. This process entails issuing search queries, extraction from new sources and reconciliation of extracted values, which are repeated until sufficient evidence is collected. We approach the problem using a reinforcement learning framework where our model learns to select optimal actions based on contextual information. We employ a deep Qnetwork, trained to optimize a reward function that reflects extraction accuracy while penalizing extra effort. Our experiments on two databases – of shooting incidents, and food adulteration cases – demonstrate that our system significantly outperforms traditional extractors and a competitive meta-classifier baseline.

Categories: Science

Astrophotographer Visits 'Alien World' on Earth in Spectacular Photo

Space.com - 14 hours 22 min ago
When astrophotographer Jeff Dai set out to see the stars shining over giant Himalayan glaciers, he stumbled upon a scene that looked more like an alien planet than an earthly landscape.
Categories: Science

WATCH LIVE @ 8:26 am ET Friday: Japanese Freighter Launching To Space Station

Space.com - 14 hours 46 min ago
Japan's robotic H-II Transfer Vehicle-6 is scheduled to launch Friday (Dec. 9) at 8:26 a.m. EST (1326 GMT) from Tanegashima Space Center. You can watch the liftoff live here at Space.com, courtesy of NASA TV, beginning at 8:00 a.m. EST (1300 GMT).
Categories: Science

Mars' Mawrth Vallis - Fly Over Where Water Once Flowed | Video

Space.com - 15 hours 2 min ago
The outflow channel is ~373 miles long (600 km) and about 1.2 miles deep 2 km). The indicators of liquid water in the past have made this area a landing site candidate for the ExoMars 2020 mission.
Categories: Science

Earth's Day Lengthens By Two Milliseconds a Century, Astronomers Find

Slashdot - 15 hours 11 min ago
Researchers at Durham University and the UK's Nautical Almanac Office compiled nearly 3,000 years of celestial records and found that with every passing century, the day on Earth lengthens by two milliseconds as the planet's rotation gradually winds down. The Guardian reports: The split second gained since the first world war may not seem much, but the time it takes for a sunbeam to travel 600km towards Earth can cost an Olympic gold medal, as the American Tim McKee found out when he lost to Sweden's Gunnar Larsson in 1972. For those holding out for a whole extra hour a day, be prepared for a long wait. Barring any change in the rate of slowing down, an Earth day will not last 25 hours for about two million centuries more. Researchers at Durham University and the UK's Nautical Almanac Office gathered historical accounts of eclipses and other celestial events from 720BC to 2015. The oldest records came from Babylonian clay tablets written in cuneiform, with more added from ancient Greek texts, such as Ptolemy's 2nd century Almagest, and scripts from China, medieval Europe and the Arab dominions. The ancient records captured the times and places that people witnessed various stages of solar and lunar eclipses, while documents from 1600AD onwards described lunar occultations, when the moon passed in front of particular stars and blocked them from view. To find out how the Earth's rotation has varied over the 2,735-year-long period, the researchers compared the historical records with a computer model that calculated where and when people would have seen past events if Earth's spin had remained constant. The astronomers found that Earth's spin would have slowed down even more had it not been for a counteracting process. Since the end of the most recent ice age, land masses that were once buried under slabs of frozen water have been unloaded and sprung back into place. The shift caused the Earth to be less oblate -- or squished -- on its axis. And just as a spinning ice skater speeds up when she pulls in her arms, so the Earth spins faster when its poles are less compressed. Changes in the world's sea levels and electromagnetic forces between Earth's core and its rocky mantle had effects on Earth's spin too, according to the scientists' report in Proceedings of the Royal Society.

Read more of this story at Slashdot.

Categories: Science

How OSIRIS-Rex Will Photograph Asteroid Bennu | Video

Space.com - 15 hours 19 min ago
The OSIRIS-REx Camera Suite (OCAMS) includes a reflecting telescope, a medium range color camera, and a wide-angle camera. The NASA mission is currently en route to the asteroid after its September 2016 launch.
Categories: Science

Paris Makes All Public Transportation Free In Battle Against 'Worst Air Pollution For 10 Years'

Slashdot - 18 hours 41 min ago
Paris has barred some cars from its streets and has made public transportation free as it suffers from the worst and most prolonged winter pollution for at least 10 years, the Airparif agency said on Wednesday. The Independent reports: Authorities have said only drivers with odd-numbered registration plates can drive in the capital region on Wednesday. Drivers of even-numbered cars were given the same opportunity on Tuesday, but could now be fined up to 35 EUR if they are caught behind the wheel. More than 1,700 motorists were fined for violations on Tuesday. Paris mayor Anne Hidalgo said images of smog blanketing the capital were proof of the need to reduce vehicle use in the city center. The air pollution peak is due to the combination of emissions from vehicles and from domestic wood fires as well as near windless conditions which means pollutants have not been dispersed, the Airparif agency said. "This is a record period (of pollution) for the last 10 years," Karine Leger of AirParif told AFP by telephone. For more than a week, Airparif has published readings of PM10 at more than 80 micrograms per cubic meter of air particles, triggering the pollution alert. Along with odd-numbered cars, hybrid or electric vehicles as well as those carrying three or more people will be allowed to roam the roads. Foreign and emergency vehicles will be unaffected.

Read more of this story at Slashdot.

Categories: Science

Adobe Flash Responsible For Six of the Top 10 Bugs Used By Exploit Kits In 2016

Slashdot - 20 hours 6 min ago
Trailrunner7 quotes a report from On the Wire: Vulnerabilities in Flash and Internet Explorer dominated the exploit kit landscape in the last year, with a high-profile bug in Flash being found in seven separate kits, new research shows. Exploit kits have long been a key tool in the arsenal of many attackers, from low-level gangs to highly organized cybercrime crews. Their attraction stems from their ease of use and the ability for attackers to add exploits for new vulnerabilities as needed. While there are dozens of exploit kits available, a handful of them attract the most use and attention, including Angler, Neutrino, Nuclear, and Rig. Researchers at Recorded Future looked at more than 140 exploit kits and analyzed which exploits appeared in the most kits in the last year, and it's no surprise that Flash and IE exploits dominated the landscape. Six of the top 10 most-refquently targeted vulnerabilities in the last year were in Flash, while the other four were in Microsoft products, including IE, Windows, and Silverlight. Flash has been a favorite target for attackers for a long time, for two main reasons: it's deployed on hundreds of millions of machines, and it has plenty of vulnerabilities. Recorded Future's analysis shows that trend is continuing, and one Flash bug disclosed October 2015 was incorporated into seven individual exploit kits. The flaw was used by a number of high-level attackers, including some APT groups. "Adobe Flash Player's CVE-2015-7645, number 10 in terms of references to exploit kits, stands out as the vulnerability with the most adoption by exploit kits. Exploit kits adopting the Adobe bug in the past year include Neutrino, Angler, Magnitude, RIG, Nuclear Pack, Spartan, and Hunter," the analysis by Recorded Future says.

Read more of this story at Slashdot.

Categories: Science

Apple Is In Talks With Hollywood For Early Access To Movies On iTunes: Bloomberg

Slashdot - 20 hours 46 min ago
Apple is talking with Hollywood studios to try and get iTunes rentals of movies that are still playing on the big screen. According to a report from Bloomberg, "some studio executives have been pushing to allow home rentals as early as two weeks after theatrical debuts and are considering a deal with iTunes as one option." Bloomberg reports: The most recent talks are part of longer-running efforts by Cupertino, California-based Apple to get new movies sooner, two of the people said. Such an arrangement could help iTunes stand out in a crowded online market for movies, TV shows and music. While the iTunes store helped Apple build a dominant role in music retailing, the company hasn't carved out a similar role in music and video streaming. Hollywood studios typically give theaters exclusive rights to new movies for 90 days or more before issuing them on DVD or making them available for online purchase. One of the concerns about iTunes is whether it will be a secure platform for delivering movies that are still in theaters, the people said. While Apple encrypts iTunes video files so they can't easily be duplicated, it's possible to use a camera to record a movie playing on a TV screen. A leak of picture that's still in theaters would jeopardize returns for the studios and cinema owners.

Read more of this story at Slashdot.

Categories: Science

SpaceX Says It’s Ready for Liftoff Again. The FAA Begs to Differ

Wired News - 21 hours 15 min ago
An Federal Aviation Administration spokesman says the agency has not given SpaceX license to launch in early January. The post SpaceX Says It's Ready for Liftoff Again. The FAA Begs to Differ appeared first on WIRED.
Categories: Science

NSA, GCHQ Have Been Intercepting In-Flight Mobile Calls For Years

Slashdot - 21 hours 26 min ago
An anonymous reader quotes a report from Reuters: American and British spies have since 2005 been working on intercepting phone calls and data transfers made from aircraft, France's Le Monde newspaper reported on Wednesday, citing documents from former U.S. spy agency contractor Edward Snowden. According to the report, also carried by the investigative website The Intercept, Air France was targeted early on in the projects undertaken by the U.S. National Security Agency (NSA) and its British counterpart, GCHQ, after the airline conducted a test of phone communication based on the second-generation GSM standard in 2007. That test was done before the ability to use phones aboard aircraft became widespread. "What do the President of Pakistan, a cigar smuggler, an arms dealer, a counterterrorism target, and a combatting proliferation target have in common? They all used their everyday GSM phone during a flight," the reports cited one NSA document from 2010 as saying. In a separate internal document from a year earlier, the NSA reported that 100,000 people had already used their mobile phones in flight as of February 2009, a doubling in the space of two months. According to Le Monde, the NSA attributed the increase to "more planes equipped with in-flight GSM capability, less fear that a plane will crash due to making/receiving a call, not as expensive as people thought." Le Monde and The Intercept also said that, in an internal presentation in 2012, GCHQ had disclosed a program called "Southwinds," which was used to gather all the cellular activity, voice communication, data, metadata and content of calls made on board commercial aircraft.

Read more of this story at Slashdot.

Categories: Science

Nintendo Offers Up To $20,000 To Hack the 3DS

Slashdot - 22 hours 6 min ago
Mickeycaskill writes: Nintendo will pay up to $20,000 for system and software vulnerabilities in the Nintendo 3DS family of handheld gaming consoles. The company is looking to prevent activities such as piracy, cheating and the circulation of inappropriate content to children. The stated goal is to "provide a secure environment for our customers so that they can enjoy our games and services. In order to achieve this goal, Nintendo is interested in receiving vulnerability information that researchers may discover regarding Nintendo's platforms." Silicon.co.uk reports: "Rewards will range from $100 to $20,000, with one given per 'qualifying piece of vulnerability information.' Hackers looking to claim a reward will have to provide Nintendo with either a proof-of-concept or a piece of functional exploit code in order to qualify."

Read more of this story at Slashdot.

Categories: Science

How your parenting tactics influence your teen's problem behaviors

Science Daily - Wed, 07/12/2016 - 11:46pm
When teens viewed their parents' parenting tactics more negatively than parents did, they showed elevated levels of aggressive behaviors, concludes a new study..
Categories: Science

Decoding cement's shape promises greener concrete

Science Daily - Wed, 07/12/2016 - 11:46pm
Materials scientists develop techniques to control the microscopic shape of cement particles for the bottom-up manufacture of stronger, more durable and more environmentally friendly concrete.
Categories: Science

Machine learning enables predictive modeling of 2-D materials

Science Daily - Wed, 07/12/2016 - 11:46pm
Scientists have used machine learning tools to create the first atomic-level model that accurately predicts the thermal properties of stanene, a 2-D material made up of a one-atom-thick sheet of tin.
Categories: Science

Qualcomm Debuts 10nm Server Chip To Attack Intel Server Stronghold

Slashdot - Wed, 07/12/2016 - 11:20pm
An anonymous reader quotes a report from Tom's Hardware: Qualcomm and its Qualcomm Datacenter Technologies subsidiary announced today that the company has already begun sampling its first 10nm server processor. The Centriq 2400 is the second generation of Qualcomm server SOCs, but it is the first in its new family of 10nm FinFET processors. The Centriq 2400 features up to 48 custom Qualcomm ARMv8-compliant Falkor cores and comes a little over a year after Qualcomm began developing its first-generation Centriq processors. Qualcomm's introduction of a 10nm server chip while Intel is still refining its 14nm process appears to be a clear shot across Intel's bow--due not only to the smaller process, but also its sudden lead in core count. Intel's latest 14nm E7 Broadwell processors top out at 24 cores. Qualcomm isn't releasing more information, such as clock speeds or performance specifications, which would help to quantify the benefit of its increased core count. The server market commands the highest margins, which is certainly attractive for the mobile-centric Qualcomm, which found its success in the relatively low-margin smartphone segment. However, Intel has a commanding lead in the data center with more than a 99% share of the world's server sockets, and penetrating the segment requires considerable time, investment, and ecosystem development. Qualcomm unveiled at least a small portion of its development efforts by demonstrating Apache Spark and Hadoop on Linux and Java running on the Centriq 2400 processor. The company also notes that Falkor is SBSA compliant, which means that it is compatible with any software that runs on an ARMv8-compliant server platform.

Read more of this story at Slashdot.

Categories: Science

The Election Is Over. The Probe Into Russian Hacks Shouldn’t Be

Wired News - Wed, 07/12/2016 - 10:51pm
Two Congressmen are calling for a commission to investigate Russian political hacking. Cybersecurity experts are backing the idea. The post The Election Is Over. The Probe Into Russian Hacks Shouldn’t Be appeared first on WIRED.
Categories: Science

Immerse Yourself in the Horror That Is an African Grassland at Night

Wired News - Wed, 07/12/2016 - 10:41pm
Listen to the sounds of lions grumbling and hyenas cackling and elephants trampling, and appreciate that humans have removed themselves from the food chain. The post Immerse Yourself in the Horror That Is an African Grassland at Night appeared first on WIRED.
Categories: Science

Google, HTC, Oculus, Samsung, Sony Join Forces To Create Global VR Association

Slashdot - Wed, 07/12/2016 - 10:40pm
Google, HTC, Oculus, Samsung, Sony and Acer have teamed up to form the Global Virtual Reality Association (GVRA) in an effort to reduce fragmentation and failure in the industry. GVRA aims to "unlock and maximize VR's potential," but there are little details as to what this may mean for consumers. TechCrunch reports: What many in the VR community have been thirsting for is some unification of standards in terms of software and hardware. Games bought in the Oculus store don't play on the Vive or PS VR. Sensors for the Vive don't work on Oculus. Sony doesn't play nice with anyone else's standards etc. etc. Valve, which makes the Steam store and SteamVR platform for the HTC Vive and others, is notably not a member of this collective so any hopes of a unified standard (like its OpenVR platform) emerging from this collective is likely not in the cards. From the GVRA press release: "The goal of the Global Virtual Reality Association is to promote responsible development and adoption of VR globally. The association's members will develop and share best practices, conduct research, and bring the international VR community together as the technology progresses. The group will also serve a resource for consumers, policymakers, and industry interested in VR."

Read more of this story at Slashdot.

Categories: Science