Categories
The Verge

T-Mobile expands its faster midband 5G network, nearly doubling its coverage

T-Mobile has announced another expansion of its 2.5GHz midband 5G network, which is now available in dozens of new cities, nearly doubling the coverage of its last major midband rollout from the end of September. With today’s expansion announcement, T-Mobile says that it now has midband 5G support in nearly 410 cities and towns in the United States. The new expansion comes just after the launch of Apple’s new 5G-compatible iPhone 12 lineup, which is expected to vastly increase the number of 5G devices in the US. T-Mobile’s 5G network combines its widely available 600MHz low-band network (which offers coverage on a nationwide scale but not much in the way of speed improvements of LTE) with the faster 2.5GHz network (which it acquired from Sprint) along with its ultra-fast mmWave network (which has the best speeds but the worst range). The actually fast flavor of T-Mobile’s 5G It’s what the company likes to call its “layer cake” approach to 5G, with the 2.5GHz chunk of the cake the “Goldilocks” portion: fast enough to offer a meaningful speed increases over regular LTE and low-band 5G but still able to be transmitted over broad chunks of cities and towns without worrying about interference from trees, walls, or buildings (problems that greatly limit mmWave). T-Mobile already has wide nationwide coverage on its low-band network, thanks to its initial 5G launch last year. The goal of the current midband expansion is to bring actual next-generation speeds to more users in a way that the…Continue readingT-Mobile expands its faster midband 5G network, nearly doubling its coverage

Categories
VentureBeat

Google’s Transporter Network uses minimal data to teach robots to stack blocks

Researchers at Google say they’ve developed an AI model architecture — Transporter Network — that enables object-grasping robots to reason about which visual cues are important and how they should be rearranged in a scene. During experiments, the researchers say their Transporter Networks achieved “superior” efficiency on a number of tasks including stacking a pyramid of blocks, assembling kits, manipulating ropes, and pushing piles of small objects. Robot grasping is a challenge. For example, robots struggle to perform what’s called “mechanical search,” which is when they have to identify and pick up an object from within a pile of other objects. Most robots aren’t especially adaptable, and there’s a lack of sufficiently capable AI models for guiding robot pincers in mechanical search — a problem that’s come to the fore as the pandemic causes companies to consider adopting automation. The Google study coauthors say Transporter Networks don’t require any prior 3D model, pose, or class category knowledge of the objects to be manipulated, instead relying only on information contained within partial depth camera data. They’re also capable of generalizing to new objects and configurations and, for some tasks, learning from a single demonstration. In fact, on 10 unique tabletop manipulation tasks, Transporter Networks trained from scratch ostensibly attained over 90% success on most tasks with objects in new configurations using 100 expert video demonstrations of the tasks. The researchers trained Transporter Networks on datasets of demonstrations ranging in number from one demonstration to 1,000 per task. They first deployed them…Continue readingGoogle’s Transporter Network uses minimal data to teach robots to stack blocks

Categories
VentureBeat

DeepMind open-sources the FermiNet, a neural network that simulates electron behaviors

In September, Alphabet’s DeepMind published a paper in the journal Physical Review Research detailing Fermionic Neural Network (FermiNet), a new neural network architecture that’s well-suited to modeling the quantum state of large collections of electrons. The FermiNet, which DeepMind claims is one of the first demonstrations of AI for computing atomic energy, is now available in open source on GitHub — and ostensibly remains one of the most accurate methods to date. In quantum systems, particles like electrons don’t have exact locations. Their positions are instead described by a probability cloud. Representing the state of a quantum system is challenging, because probabilities have to be assigned to possible configurations of electron positions. These are encoded in the wavefunction, which assigns a positive or negative number to every configuration of electrons; the wavefunction squared gives the probability of finding the system in that configuration. The space of possible configurations is enormous — represented as a grid with 100 points along each dimension, the number of electron configurations for the silicon atom would be larger than the number of atoms in the universe. Researchers at DeepMind believed that AI could help in this regard. They surmised that, given neural networks have historically fit high-dimensional functions in artificial intelligence problems, they could be used to represent quantum wavefunctions as well. Above: Simulated electrons sampled from the FermiNet move around a bicyclobutane molecule. By way of refresher, neural networks contain neurons (mathematical functions) arranged in layers that transmit signals from input data and slowly…Continue readingDeepMind open-sources the FermiNet, a neural network that simulates electron behaviors

Categories
Mashable

Nokia wins NASA contract to put a 4G network on the moon

Soon, astronauts on moon missions won’t have any excuse for not answering their texts. NASA has awarded Nokia of America $14.1 million to deploy a cellular network on the moon. The freaking moon. The grant is part of $370 million worth of contracts signed under NASA’s “Tipping Point” selections, meant to advance research and development for space exploration.  Nokia’s plan is to build a 4G/LTE network, and eventually transition to 5G (just like the rest of us). It will be “the first LTE/4G communications system in space,” according to NASA’s announcement. “The system could support lunar surface communications at greater distances, increased speeds, and provide more reliability than current standards,” the announcement also reads. To the moon! 🌕 We are excited to have been named by @NASA as a key partner to advance “Tipping Point” technologies for the moon, to help pave the way towards sustainable human presence on the lunar surface. So, what technology can you expect to see? (1/6) pic.twitter.com/wDNwloyHdP — Bell Labs (@BellLabs) October 15, 2020 Nokia’s research arm, Bell Labs, provided more details in a Twitter thread. The company intends for the network to support wireless operation of lunar rovers and navigation, as well as streaming video.  The network is built to be compact and efficient, as well as “specially designed to withstand the extreme temperature, radiation and vacuum conditions of space.” According to UPI, NASA said in a live broadcast of the announcement that the network would extend to spacecraft, and help develop technology fit…Continue readingNokia wins NASA contract to put a 4G network on the moon

Categories
TechCrunch

Hands on with Telepath, the social network taking aim at abuse, fake news and, to some extent, ‘free speech’ – TechCrunch

There’s no doubt that modern social networks have let us down. Filled with hate speech and abuse, moderation and anti-abuse tools were an afterthought they’re now trying to cram in. Meanwhile, personalization engines deliver us only what will keep us engaged, even if it’s not the truth. Today, a number of new social networks are trying to flip the old model on its head — whether that’s attempting to use audio for more personal connections, like Clubhouse, eliminate clout chasing, like Twelv, or, in the case of new social network Telepath, by designing a platform guided by rules that focus on enforcing kindness, countering abuse, and disabling the spread of fake news. Many of these early efforts are already facing challenges. Private social network Clubhouse has repeatedly demonstrated that allowing free-flowing communication in the form of audio conversations is an area that’s notoriously difficult to moderate. The app, though still unavailable to the broader public, courted controversy in September when it allowed anti-Semitic content to be discussed in one of its chat rooms. In the past, it had also allowed users to harass an NYT reporter openly. Meanwhile, Twelv, a sort of Instagram alternative, ditches the “Like” button concept and all the other features now overloading Instagram, which had once been just a photo-sharing network. But, unfortunately, this also means there’s no easy way to find and follow interesting users or trends on Twelv — you have to push friends to join the app with you or know someone’s username to…Continue readingHands on with Telepath, the social network taking aim at abuse, fake news and, to some extent, ‘free speech’ – TechCrunch

Categories
Wired

Researchers Found 55 Flaws in Apple’s Corporate Network

For months, Apple’s corporate network was at risk of hacks that could have stolen sensitive data from potentially millions of its customers and executed malicious code on their phones and computers, a security researcher said on Thursday. ARS TECHNICA This story originally appeared on Ars Technica, a trusted source for technology news, tech policy analysis, reviews, and more. Ars is owned by WIRED’s parent company, Condé Nast. Sam Curry, a 20-year-old researcher who specializes in website security, said that, in total, he and his team found 55 vulnerabilities. He rated 11 of them critical because they allowed him to take control of core Apple infrastructure and from there steal private emails, iCloud data, and other private information. The 11 critical bugs were: Remote Code Execution via Authorization and Authentication Bypass Authentication Bypass via Misconfigured Permissions allows Global Administrator Access Command Injection via Unsanitized Filename Argument Remote Code Execution via Leaked Secret and Exposed Administrator Tool Memory Leak leads to Employee and User Account Compromise allowing access to various internal applications Vertica SQL Injection via Unsanitized Input Parameter Wormable Stored XSS allows Attacker to Fully Compromise Victim iCloud Account Wormable Stored XSS allows Attacker to Fully Compromise Victim iCloud Account Full Response SSRF allows Attacker to Read Internal Source Code and Access Protected Resources Blind XSS allows Attacker to Access Internal Support Portal for Customer and Employee Issue Tracking Server-Side PhantomJS Execution allows attacker to Access Internal Resources and Retrieve AWS IAM Keys Apple promptly fixed the vulnerabilities after Curry reported…Continue readingResearchers Found 55 Flaws in Apple’s Corporate Network

Categories
Tech Radar

Watch out Cloudflare, Cisco is launching a content delivery network service

The CDN market is about to get a whole lot more competitive as Cisco, Qwilt and Digital Alpha have announced that they will release a new as-a-service offering based on Open Caching with UK telecom BT as its flagship customer. Streaming video content is increasingly being delivered in 4K and with 8K just around the corner, network capacity demands will need to be increased. At the same time, internet video traffic is expected to comprise 82 percent of all consumer internet traffic by 2022 which is up from 73 percent in 2017. Increased performance requirements are helping to accelerate the shift away from traditional content delivery models. As a result of this, an opportunity has opened for service providers to use their edge assets to deploy their own distributed CDN capabilities. Open Caching is an open architecture developed and endorsed by the Streaming Video Alliance that offers a platform which federates content delivery infrastructure inside service provider networks into a global CDN with open APIs for content publishers. CDN as-a-service The new CDN offering will use investments from Digital Alpha combined with Qwilt’s content delivery platform based on Open Caching and Cisco’s edge computer and networking infrastructure to deliver the solution as-a-service to service providers of all sizes around the world. BT has already deployed the new solution to add multiple terabits per second of capacity and to provide cost-effective, high-quality streaming video to meet the growing demand of its customers. The telecom decided to make the transition to Open…Continue readingWatch out Cloudflare, Cisco is launching a content delivery network service

Categories
Tech Radar

Three invests £100m to boost 4G and 5G network in 250 towns and cities

Three is partnering with SEE Enterprise Telecoms on a £100 million infrastructure investment programme that will boost the mobile operator’s 4G and 4G services. No details about specific upgrades have been disclosed other than that the network will increase capacity and reliability across 250 towns and cities across the UK. SSE has been a key partner for Three in its 5G rollout, connecting Three’s 20 core UK data centres to fibre and unbundling BT exchanges to boost backhaul (the connection between a site and the core network). Three backhaul Fully unbundled lines allow service providers to use their own equipment at telephone exchanges, giving them greater control over the connection. This means SSE has greater flexibility in how it connects Three’s sites to the wider network. An expansion of this programme will see SSE unbundle a further 450 exchanges, extending SSE’s fibre footprint from 20,000 km to 30,000 by the middle of 2021. This will allow for further enhancements in Three’s mobile service in the future. Three uses a combination of suppliers, including BT, SSE Enterprise Telecoms and Virgin Media and began working with CityFibre to provide backhaul to its 5G sites outside of London earlier this year. In September it expanded the agreement with CityFibre to cover a further 1,300 masts in 59 towns and cities. “We are investing significantly to transform our entire network experience which will support our customers who use three times more data than the average UK consumer,” said Susan Buttsworth, Three UK COO. “SSE…Continue readingThree invests £100m to boost 4G and 5G network in 250 towns and cities

Categories
The Next Web

Facebook removes Chinese network of fake accounts that used AI-generated faces

Facebook has taken down a China-based network of fake accounts that used AI-generated faces to spread government propaganda across the platform, according to research by analytics firm Graphika. The social network announced on Tuesday that it had removed 155 accounts, 11 Pages, nine Groups, and six Instagram accounts for “coordinated inauthentic behavior on behalf of a foreign or government entity.” The fake accounts predominantly focused on Southeast Asia, where they posted about events including Beijing’s interests in the South China Sea and developments in Hong Kong. But a smaller cluster posing as Americans posted content that both supported and criticized then-presidential candidates Pete Buttigieg, Joe Biden, and Donald Trump. Before the takedown, Facebook asked Graphika to analyze the data. The Pentagon-linked firm found that some accounts had stolen their profile photos from real people, which can be exposed as inauthentic through reverse image searches. Others tried to evade detection by using Generative Adversarial Networks (GAN) to create fake pictures, a tactic that Graphika says “has exploded in the last year.” [Read: Are EVs too expensive? Here are 5 common myths, debunked] The company found 12 profile pictures that it suspects were AI-generated, due to their distorted backgrounds and asymmetrical peripheral features such as ears, glasses, and hair. Credit: GraphikaNine superimposed profile pictures show the alignment of their eyes, “an indicator of synthetic image generation,” according to Graphika. Graphika spotted some of these signals by rendering the images opaque and then superimposing them on top of each other to expose the alignment of the features. The firm…Continue readingFacebook removes Chinese network of fake accounts that used AI-generated faces

Categories
TechCrunch

Chinese propaganda network on Facebook used AI-generated faces – TechCrunch

Facebook removed two networks of fake accounts spreading government propaganda on the platform Tuesday, one originating in China and one in the Philippines. In its latest report on this kind of coordinated campaign, the company says it took down 155 Facebook accounts, 11 pages, 9 groups and 7 Instagram accounts connected to the Chinese activity and 57 accounts, 31 Pages and 20 Instagram accounts for the activity in the Philippines. Both operations broke Facebook’s rules against “coordinated inauthentic behavior on behalf of a foreign or government entity.” The company released the report Thursday in coordination with Graphika, a social analytics company that specialized in disinformation. Graphika regularly analyzes this kind of activity in coordination with Facebook and its reports dive into more depth about techniques. In a sign of the times, Graphika found that the Chinese network of fake accounts employed faces created through an AI technique known as GANs (Generative Adversarial Networks). Those fake faces are employed to elude detection, but because their visual signature often ends up with subtle quirks and anomalies, GANs can sometimes be easily detected. “This form of AI is readily available online, and its use (or abuse) by covert operations has exploded in the last year,” according to Graphika’s report, which identified a dozen GAN-generated images from the Chinese information operation. “A year ago, this was a novelty,” Graphika’s Ben Nimmo wrote on Twitter. “Now it feels like every operation we analyse tries this at least once.” GANs examples via Graphika The Chinese campaign,…Continue readingChinese propaganda network on Facebook used AI-generated faces – TechCrunch