Stories
Slash Boxes
Comments

SoylentNews is people

Log In

Log In

Create Account  |  Retrieve Password


Site News

Join our Folding@Home team:
Main F@H site
Our team page


Funding Goal
For 6-month period:
2022-07-01 to 2022-12-31
(All amounts are estimated)
Base Goal:
$3500.00

Currently:
$438.92

12.5%

Covers transactions:
2022-07-02 10:17:28 ..
2022-10-05 12:33:58 UTC
(SPIDs: [1838..1866])
Last Update:
2022-10-05 14:04:11 UTC --fnord666

Support us: Subscribe Here
and buy SoylentNews Swag


We always have a place for talented people, visit the Get Involved section on the wiki to see how you can make SoylentNews better.

Best movie second sequel:

  • The Empire Strikes Back
  • Rocky II
  • The Godfather, Part II
  • Jaws 2
  • Star Trek II: The Wrath of Khan
  • Superman II
  • Godzilla Raids Again
  • Other (please specify in comments)

[ Results | Polls ]
Comments:90 | Votes:153

posted by takyon on Tuesday May 29 2018, @10:29PM   Printer-friendly
from the crisp-fakes dept.

There has been some controversy over Deepfakes, a process of substituting faces in video. Almost immediately, it was used for pornography. While celebrities were generally unamused, porn stars were alarmed by the further commodification of their rôle. The algorithm is widely available and several web sites removed objectionable examples. You know something is controversial when porn sites remove it. Reddit was central for Deepfakes/FakeApp tech support and took drastic action to remove discussion after it started to become synonymous with fictitious revenge porn and other variants of anti-social practices.

I found a good description of the deepfakes algorithm. It runs via a standard neural network library but requires considerable processing power on specific GPUs. I will describe the video input (with face to be removed) as the source and the face to be replaced as the target. The neural network is trained with the target face only. The source is distorted and the neural network is trained to approximate reference images of the target. When the neural network is given the source, it has been trained to "undistort" the source to target.

[Continues...]

If there are multiple faces in a frame of video, face recognition restricts input to the most likely face. Indeed, for maximum efficiency, this technique is used to crop source video in all cases. The trick that makes the process feasible is that the neural network is only trained with the target face. Furthermore, given the use of libraries, the unique code to achieve this objective is shockingly small.

A friend attempted to mix DeepFakes with the Internet meme of Downfall parodies. There is an infamous scene in the film Downfall (not to be confused with the film Falling Down) where Adolf Hitler rants prior to defeat. Unfaithful subtitles of the German dialog have been used to parody everything from corporate sales targets to sportsball management to the ongoing medical abuse of transsexual patients. Until now, only the words in the subtitles changed. The audio and video was otherwise unchanged. My friend hoped that it would be possible to insert the likeness of people being parodied.

Unfortunately, it doesn't work with the current algorithm. The number of faces is not a problem. The clipping and occlusion prevents the neural network from working effectively. It should be possible with an extension of the current algorithm but it is currently impractical.

A further development, found by the same friend, is the automatic conversion of a one sentence description into a very short video. The example system uses Flintstones cartoons. An example sentence would be "Fred dancing in the kitchen" and a rough but valid video is created which matches the description. Potentially, it would be possible to automatically convert a novel into a 100 minute film with no human intervention. Given that novels are frequently converted into films, there is a large amount of example data which may be used as reference. I know this would only be moderately easier than making a holodeck but experts may not be aware of the progress towards either goal.

takyon: An algorithm can also be used to manipulate facial movements to match video or audio input (see this example of Jordan Peele controlling Barack Obama's face). DARPA is holding an event that will task experts with making and catching "deepfakes".

Researchers have also created short "movies" (64x64, 32-frame animated GIFs) from text descriptions. It may be possible to synthesize scenes for a full length movie in the future without needing strong AI. After all, procedural generation could be used to create and populate a virtual city (like the one in Big Hero 6), and then it's a matter of writing some kind of coherent narrative and "shooting" it. A "director neural network" could be trained to mimic the cinematography techniques of films created by humans, and then apply the results to the virtual environment.


Original Submission

posted by janrinok on Tuesday May 29 2018, @09:08PM   Printer-friendly
from the light-relief dept.

Battling demons and normal guns don't work? Here's help in the form of this mystical gun that channels the arcane powers of the mystic realm before firing off a blast to blow evil away.

Made by Japanese steampunk designer FriskP, the gun delivers an impressive power-up animation that doesn't require the use of post-processing computer generated effects. Instead, it relies on the Phantom, an LED fan made by a tech startup called Life is Style. The fan is capable of projecting high resolution images in the air as the fan spins, and is also programmable.

This has led to other cool effects, such as the hand spells from Marvel's Doctor Strange as seen in the video. If you're thinking of getting one for your next cosplay project, well, you'll have to wait, as pricing for the Phantom has yet to be revealed. Don't expect it to be cheap though.


Original Submission

posted by janrinok on Tuesday May 29 2018, @07:40PM   Printer-friendly
from the generational-attitude-shift dept.

Al Jazeera reports

Polls have closed in an Irish referendum on abortion that could represent a change in the path of a country that was once one of Europe's more socially conservative.

Voters turned out in large numbers on [May 25] to have their say on whether to repeal the country's Eighth Amendment, which outlaws abortion by giving equal rights to the unborn.

An exit poll, conducted for the Irish Times by Ipsos/MRBI, suggested that the country voted by a landslide margin to change the constitution so that abortion can be legalised.

The vote to repeal the constitutional ban was predicted to win by 68 percent to 32 percent, according to the poll of 4,000 voters, the Irish Times said.

[...] If the proposal to repeal the Eighth Amendment is defeated on [May 25], the country will not have a second referendum and it could be another 35 years before voters have their say on the matter again, [Prime Minister Leo] Varadkar said.

[...] 78 percent of the Irish population is Catholic

[...] Thousands of people living abroad returned home to vote. Ireland is one of the few countries in the European Union that does not allow those abroad to vote via post or in embassies.

Those away for less than 18 months remain eligible to vote at their former local polling station. Those living on the Atlantic islands cast their ballot a day early to help prevent delays in transportation and counting the ballot papers.

When the constitutional amendment to instate the ban was voted on in 1983, 66.9 percent voted "yes," and 33.1 percent voted "no".

Widely reported, including:


Original Submission

posted by janrinok on Tuesday May 29 2018, @06:23PM   Printer-friendly
from the for-true-hackers dept.

Glenn Grant has blogged recently about going down the USB Reverse Engineering rabbit hole. He does a deep dive into the software and hardware used to reverse engineer, do protocol analysis, do hardware hacking, and do what whatever else would be involved in implementing custom drivers for arbitrary hardware. USB is a ubquitous, industry standard for cables, connectors, and their supporting protocols with a surprising amount of computing power running internally on chips inside the USB devices themselves.


Original Submission

posted by janrinok on Tuesday May 29 2018, @04:53PM   Printer-friendly
from the jobs-for-everyone-else dept.

Trump Administration Makes 15,000 Additional H-2B Visas available

https://www.marketwatch.com/story/us-to-allow-another-15000-unskilled-foreign-workers-on-temporary-h-2b-visas-2018-05-26

The Department of Homeland Security said Friday it would provide businesses another 15,000 H-2B visas to bring low-skilled foreign workers to the U.S. this summer, offering a modest infusion to the popular program.

The number of visas available each year for seasonal work is capped by statute at 66,000, evenly divided between the summer and winter seasons. Congress declined to lift that cap during negotiations this spring. It did, however, give the secretary of homeland security authority to issue up to 69,000 more this summer if she determines there is sufficient need.

A range of businesses—including fisheries, landscapers and those in summer tourist spots—have complained about worker shortages and have been waiting to see if Secretary Kirstjen Nielsen will use that authority. But people who support restrictions on immigration, including some in the White House, argue that foreign workers drive down American wages and oppose additional visas.

Faced with a similar choice last summer, then-DHS Secretary John Kelly also provided an additional 15,000 visas but cast it as a one-time only move.

Also reported at:


Original Submission

posted by janrinok on Tuesday May 29 2018, @03:12PM   Printer-friendly
from the recognition-for-the-right-reasons dept.

Submitted via IRC for Runaway1956

This year's Ms Geek Africa is Salissou Hassane Latifa, 21, from Niger. Her winning design is an app that helps communication between people caring for accident victims and the emergency services, and allows medical staff to advise on basic first aid before they arrive at the scene.

"Ms Geek has already changed the perception of what girls can do," says Esther Kunda of the Next Einstein Forum, a founding member of competition organiser Girls in ICT Rwanda.

The contest was set up as part of a nationwide effort to transform Rwanda from a small agricultural economy into an engine of technological innovation, with women and girls at the forefront of the revolution.

The government has set a target of achieving gender parity in the information communications technology sector by 2020, an ambitious goal in a worldwide industry notorious for its lack of diversity. But through educational campaigns, scholarships and mentorship programmes, Rwanda is determined to become a global leader for women in ICT.

"It's a good place to be a woman in tech right now," Kunda says of Rwanda.

Source: https://www.theguardian.com/global-development/2018/may/28/brilliance-overtakes-beauty-ms-geek-africa-spotlights-tech-genius-salissou-hassane-latifa


Original Submission

posted by janrinok on Tuesday May 29 2018, @01:46PM   Printer-friendly
from the land-of-the-watched dept.

At the Private Internet Access Blog, Glyn Moody writes how Amazon and US schools are following in China's footsteps to normalize automatic facial recognition and constant surveillance. Materials gained Freedom of Information Act requests by the ACLU have documented that Amazon has been marketing in its hosted "Rekognition" products to both police forces and schools to facilitate mass surveillance inside the US and to inure the coming generations to it.

Amazon has developed a powerful cloud-based facial recognition system called "Rekognition", which has major implications for privacy. It is already being used by multiple US police forces to carry out surveillance and make arrests, the ACLU has learned.

Amazon claims that Rekognition offers real-time face matching across tens of millions of individuals held in a database, and can detect up to 100 faces in a single photo of a crowd. Rekognition can be used to analyze videos, and to track people even when their faces are not visible, or as they go in and out of the scene.

As a result of these disclosures, a coalition of organizations including the ACLU has sent a letter to Amazon's CEO Jeff Bezos demanding that the company stop providing its facial recognition tool to the government. The ACLU has also launched a petition that calls for the same.

Emails obtained through freedom of information requests submitted by the ACLU show that Amazon has worked with the city of Orlando, Florida, and the Washington County Sheriff's Office in Oregon to roll out Rekognition in those locations. In addition, law enforcement agencies in California, Arizona, and multiple domestic surveillance "fusion centers" have indicated interest in Rekognition, although it is not clear how many of these have gone on to deploy the system. Orlando has used Rekognition to search for people in footage drawn from the city's video surveillance cameras. Washington County, meanwhile, has built a Rekognition-based mobile app that its deputies can use to run any image against the county's database of 300,000 faces.


Original Submission

posted by janrinok on Tuesday May 29 2018, @12:54PM   Printer-friendly
from the again dept.

A man has shot dead two police officers and a civilian in the eastern Belgian city of Liège.

The gunman took a female cleaner hostage at a school before being killed by police. Two other police officers were also injured.

The man's motive is not yet clear but the incident is being treated as terrorism.

Police sources quoted in local media said the man was heard shouting "Allahu Akbar" ("God is greatest" in Arabic).

Belgian broadcaster RTBF said the gunman was let out from prison on temporary release on Monday where he had been serving time on drug offenses. It said that he may have been radicalised while in jail.

The shooting unfolded late morning on Tuesday near a cafe in the city centre.

Update: 16:56 UTC

More recent reporting states:

https://edition.cnn.com/2018/05/29/europe/liege-shooting-intl/index.html

The incident occurred at around 10:30 a.m. when an assailant stabbed two policewomen from behind, before stealing their service weapons and using them on the officers, Liege Prosecutor Philippe Dulieu said at a news conference on Tuesday.

After killing the two officers, the attacker continued walking through the street and opened fire on a parked vehicle, fatally wounding the driver inside, Dulieu added.

https://news.sky.com/story/belgian-police-launch-terror-probe-after-shooting-of-police-and-bystander-11388883

The gunman also killed a 22-year-old male car passenger on the Boulevard d'Avory, before taking a female cleaner hostage at a nearby high school.

She was released when police shot dead the attacker, who has been named by local media as Belgian national Benjamin Herman.

See also, thanks to C0lo:


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @12:15PM   Printer-friendly
from the pushing-boarders dept.

Possibly the first academic conference on skateboarding starts on Fri 1 Jun 2018 with an introduction to skateboarding and free beer (if you're old enough and I'm not sure mixing beer and skateboards is a good idea). The call for papers finished in Mar 2018 but the speakers are a mostly a mix of former professional skateboarders running sportwear brands and amateur skateboards who are now academics.

The print edition of the Evening Standard newspaper described it as a board meeting. That's an old joke and I believe that referring to surfboarding and skateboarding as board meetings was in a promotional video from the California Tourist Board. (Presumably inspired during a board meeting.)


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @10:43AM   Printer-friendly
from the trillium-and-cambricon-and-cavium,-oh-my! dept.

ARM Details "Project Trillium" Machine Learning Processor Architecture

[ARM has detailed] more of the architecture of what Arm now seems to more consistently call their "machine learning processor" or MLP from here on now. The MLP IP started off a blank sheet in terms of architecture implementation and the team consists of engineers pulled off from the CPU and GPU teams.

With the MLP Arm set out to provide three key aspects that are demanded in machine learning IPs: Efficiency of convolutional computations, efficient data movement, and sufficient programmability. From a high level perspective the MLP seems no different than many other neural network accelerator IPs out there. It still has a set of MAC engines for the raw computational power, while offering some sort of programmable control flow block alongside a sufficiently robust memory subsystem.

Cambricon, Makers of Huawei's Kirin NPU IP, Build A Big AI Chip and PCIe Card

Cambricon Technologies, the company in collaboration with HiSilicon / Huawei for licensing specialist AI silicon intellectual property for the Kirin 970 smartphone chipset, have gone solo and created their own series of chips for the data center. The IP inside the Kirin 970 is known as Cambricon-1A, the company's first licensable IP. At the time, finding information on Cambricon was difficult: its website was a series of static images with Chinese embedded into the image itself. Funnily enough, we used the AI-accelerated translate feature on the Huawei Mate 10 to translate what the website said. Fast forward 12-18 months, and the Cambricon website is now interactive and has information about upcoming products. A few of which were announced recently.

Built on TSMC's 16FF, the MLU-100 is an 80W chip with a capability of 64 TFLOPS of traditional half-precision or 128 TOPS using the 8-bit integer metric commonly used in machine learning algorithms. This is at 1.0 GHz, or the 'standard' mode – Cambricon's CEO, Dr Chan Tianshi, stated that their new chip has a high-performance mode at 1.30 GHz, which allows for 83.2 TFLOPS (16-bit float) or 166.4 TOPS (8-bit int) but rises to 110W. This technically decreases performance efficiency, but allows for a faster chip. All this data relies on sparse data modes being enabled.

[...] David Schor from WikiChip (the main source of this article) states that this could be NVIDIA's first major ASIC competition for machine learning, if made available to commercial partners. To that end, Cambricon is also manufacturing a PCIe card.

Assessing Cavium's ThunderX2: The Arm Server Dream Realized At Last

A little less than 2 years ago, we investigated the first Arm server SoC that had a chance to compete with midrange Xeon E5s: the Cavium ThunderX. The SoC showed promise, however the low single-threaded performance and some power management issues relegated the 48-core SoC to more niche markets such as CDN and Web caching. In the end, Cavium's first server SoC was not a real threat to Intel's Xeon.

But Cavium did not give up, and rightfully so: the server market is more attractive than ever. Intel's datacenter group is good for about 20 Billion USD (!) in revenue per year. And even better, profit margins are in 50% range. When you want to profits and cash flow, the server market far outpaces any other hardware market. So following the launch of the ThunderX, Cavium promised to bring out a second iteration: better power management, better single thread performance and even more cores (54).


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @09:11AM   Printer-friendly
from the malicious-software dept.

A French Free Software organization, April, has announced that a German Documentary from the ARD, "The Microsoft Cyber Attack" has been released in English thanks to Deutsche Welle (DW). It is an informative and objective film about the inappropriate relations between a certain infamous corporation and the various public administrations. The documentary first aired on February 19th, 2018 by the German public broadcaster (ARD).

In May 2017, hundreds of thousands of computers running Microsoft Windows operating systems were disabled by the WannaCry cyber attack. How could a single malware program simultaneously cripple companies, hospitals and even government intelligence services all around the globe? Microsoft Windows software programs proved to be their common Achilles heel. Companies and private individuals use software from Microsoft. Government and public administrations from Helsinki to Lisbon run it, too. That makes all of them vulnerable to attacks from hackers and spies. Microsoft Window's dominance also undermines European procurement legislation, impedes technological progress and costs Europe a bundle. Journalist Harald Schumann and his team of Investigate Europe researchers have spoken with insiders and administrators from all across the continent. The German government's former IT director, Martin Schallbruch, tells us how countries are becoming increasingly dependent on Microsoft. A legal expert from the Netherlands describes how the European Commission and governments are breaking European laws regulating public tenders. Hamburg's data protection commissioner, Johannes Caspar, warns that Microsoft Windows systems expose individuals' private data to the prying eyes of US intelligence services. Internal documents show that Germany's Federal Office for Information Security (BSI) suspects this, too. The European Parliament and the German parliament have responded by repeatedly demanding that government IT systems be converted to open source software. Their source codes can be accessed freely and copied at will, which would enable European security services to use, alter and monitor them. Italy's army is going open source, as have police in France, Lithuania, and in the cities of Rome and Barcelona. Why do most governments resist the alternatives, or fall back into Microsoft's clutches, as Munich city authorities did. The EU's Commissioner for the Digital Single Market, Andrus Ansip, and other key players have the answers.

The video itself, « The Microsoft Cyber Attack », is available at Youtube and is about 43 minutes long.


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @07:39AM   Printer-friendly
from the at-what-cost dept.

Yahoo Finance reports

Poverty-alleviation programs like food stamps (SNAP), Social Security, and other "welfare" programs are broadly effective at reducing poverty, a new study from University of Chicago researchers found.

The study, performed by researchers Bruce Meyer and Derek Wu, conducted a more comprehensive analysis than most studies, because it used administrative data from the programs' payment records, not just survey data of recipients from the Census Bureau.

[...] For the elderly, Wu said the research found that Social Security benefits "single-handedly slashes poverty by 75%." Social Security's overall effect on all poverty is also enormous, responsible for by far the largest poverty reduction among all these programs, the study said.


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @06:07AM   Printer-friendly
from the flood-insurance-FTW dept.

Common Dreams reports

A Maryland city was devastated [May 27] after 6-inches of heavy rain caused a downtown flash flood. Major damage is reported and many cars have been swept away.

Ellicott City was still recovering from a flash flood two years ago that killed two and forced the historic city to rebuild much of its Main Street. Residents said Sunday's flood seemed even worse than the storm in July 2016--which was called an extremely rare "one-in-1,000 year event", and cost the city tens of millions of dollars in damages.

Additional information at:
The Baltimore Sun
The Washington Post
USAToday


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @04:35AM   Printer-friendly
from the like-a-bad-penny dept.

Submitted via IRC for guy_

Plants, like all living things, need nitrogen to build amino acids and other essential biomolecules. Although nitrogen is the most abundant element in air, the molecular form of nitrogen found there is largely unreactive. To become useful to plants, that nitrogen must first be "fixed," or busted out of its molecular form and linked with hydrogen to make ammonia. The plants can then get at it by catalyzing reactions with ammonia.

But plants can't fix nitrogen. Bacteria can.

Some legumes and a few other plants have a symbiotic relationship with certain bacterial species. The plants build specialized structures on their roots called nodules to house and feed the bacteria, which in turn fix nitrogen for the plants and assure them a steady supply of ammonia. Only 10 families of plants have the ability to do this, and even within these families, most genera opt out. Ever since the symbiosis was discovered in 1888, plant geneticists have wondered: why? If you could ensure a steady supply of nitrogen for use, why wouldn't you?

A global consortium of geneticists sequenced and compared the genomes of 37 plants—some symbiotic, some not; some that build nodules, some not; some agriculturally relevant, some not—to try to find out what was going on. The group's genetic analysis of the conundrum was reported in Science.

Source: https://arstechnica.com/science/2018/05/plants-repeatedly-got-rid-of-their-ability-to-obtain-their-own-nitrogen/


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @03:03AM   Printer-friendly
from the not-as-fragile-as-believed dept.

Intel and Micron's jointly-developed 3D QLC (4 bits per cell) NAND memory is featured in a new Micron enterprise SSD, the 5210 ION. The drive will have a capacity of 1.92 TB, 3.84 TB, or 7.68 TB, and a write endurance of less than 1 drive write per day (possibly as low as 0.1 DWPD):

The cost reduction brought by QLC NAND is a much-awaited advance for enterprise storage. Most NAND flash manufacturers have started sampling QLC NAND within the past year, generally built on the same 64-layer 3D NAND processes that current-generation TLC NAND uses. Micron has previously shown wafers of 512Gb 64-layer QLC when announcing the addition of QLC to their roadmap, but today they are also announcing a 1Tb 64L QLC part—the first 1Tb memory chip to hit commercial availability. That 1Tb part is organized as four planes that can be processing I/O commands in parallel, compared to two planes for previous Intel/Micron NAND parts. This helps offset most of the performance loss associated with increasing per-die capacity. Thanks to the "CMOS under the array" design of Intel/Micron 3D NAND, the extra peripheral circuitry requried by doubling the number of planes doesn't add much to the overall die size.

It was initially feared that QLC write endurance would be low enough that drives would need to be treated more or less as write-once, read-many (WORM) devices, requiring careful handling on the software side. With multiple manufacturers now rating their QLC NAND for around 1k P/E cycles, it is clear that QLC-based SSDs aren't too fragile and can handle many existing workloads without needing major software changes to reduce writes.

Micron is primarily marketing the 5210 ION SSDs as replacement for hard drives, rather than replacements for any existing tier of enterprise SSD products. In this role, the 5210 ION will have clear advantages in density (with 2-8TB per 2.5" drive) and performance. QLC NAND only provides incremental improvements to cost, so the 5210 ION won't be matching 7200RPM hard drives for price per GB, but 10k RPM drives will probably be feeling the pressure, especially from TCO calculations that take into account the power efficiency advantages of SSDs.

The next generation of Intel/Micron 3D NAND will have 96 layers, potentially using string-stacking to combine two 48-layer dies. After that, Intel and Micron will go their separate ways.


Original Submission

posted by Fnord666 on Tuesday May 29 2018, @01:33AM   Printer-friendly
from the then-again-what-can? dept.

The hysteria about the future of artificial intelligence (AI) is everywhere. There seems to be no shortage of sensationalist news about how AI could cure diseases, accelerate human innovation and improve human creativity. Just looking at the media headlines, you might think that we are already living in a future where AI has infiltrated every aspect of society.

While it is undeniable that AI has opened up a wealth of promising opportunities, it has also led to the emergence of a mindset that can be best described as "AI solutionism". This is the philosophy that, given enough data, machine learning algorithms can solve all of humanity's problems.

But there's a big problem with this idea. Instead of supporting AI progress, it actually jeopardises the value of machine intelligence by disregarding important AI safety principles and setting unrealistic expectations about what AI can really do for humanity.

In only a few years, the pendulum has swung from the dystopian notion that AI will destroy humanity to the utopian belief that our algorithmic saviour is here.

[...] Examples demonstrate that there is no AI solution for everything. Using AI simply for the sake of AI may not always be productive or useful. Not every problem is best addressed by applying machine intelligence to it. This is the crucial lesson for everyone aiming to boost investments in national AI programmes: all solutions come with a cost and not everything that can be automated should be.

The Conversation

What is your take on this? Do you think AI (as currently defined), can solve any of the problems, man-made and otherwise, of this world?


Original Submission

Today's News | May 30 | May 28  >