Cardiovascular disease (CVD) is the leading cause of death worldwide. About 610,000 people die of heart attacks and strokes in the U.S. every year, according to the Center for Disease Control and Prevention, and worldwide, the number stands at about 17.9 million. CVD isn’t impossible to predict, fortunately — there’s a strong risk factor in coronary artery calcium (CAC) deposits that restrict blood flow. Unfortunately, measuring CAC requires experts who can closely inspect computerized tomography (CT) scans for worsening signs and symptoms.But there’s hope yet for a more automated approach.A newly published paper on the preprint server Arxiv.org (“Direct Automatic Coronary Calcium Scoring in Cardiac and Chest CT“) proposes an artificially intelligent (AI) system that can evaluate and score CAC without human supervision. That’s not especially novel — automated CAC tests have been around for a while. However, the coauthors claim that their system is up to hundreds of times faster than state-of-the-art methods....

It’s been an eventful week in tech. Amazon announced it would abandon plans to open one of its two HQ2 locations in New York City, and the company also acquired Wi-Fi mesh network startup Eero for an undisclosed sum — a hint at Amazon’s future smart home ambitions. The California Department of Motor Vehicles released reports from companies currently testing self-driving cars — like Apple, Alphabet’s Waymo, and GM Cruise. Google pledged to spend $13 billion on U.S. datacenters and offices in 24 states this year, and driverless truck startup TuSimple raised $95 million at a $1 billion valuation, joining the ranks of Aurora and Nuro as one of the best-funded companies in the autonomous vehicle industry....

Reinforcement learning — a machine learning training technique that uses rewards to drive AI agents toward certain goals — is a reliable means of improving said agents’ decision-making, given plenty of compute, data, and time. But it’s not always practical; model-free approaches, which aim to get agents to directly predict actions from observations about their world, can take weeks of training.Model-based reinforcement learning is a viable alternative — it has agents come up with a general model of their environment they can use to plan ahead. But in order to accurately forecast actions in unfamiliar surroundings, those agents have to formulate rules from experience. Toward that end, Google in collaboration with DeepMind today introduced the Deep Planning Network (PlaNet) agent, which learns a world model from image inputs and leverages it for planning. It’s able to solve a variety of image-based tasks with up to 5000 percent the data efficiency, Google says, while maintaining competitiveness with advanced model-free agents....

Facebook has revealed additional details about the development of Portal, its first in-house video calling device, including the tidbit that early prototypes included a motor that let Portal swivel to face video subjects.If the plan was to sell lots of video calling devices, it’s probably best that a swiveling Portal was never released. Following Facebook’s persistent parade of controversy, privacy concerns raised by Portal were bad enough without the device turning to face you when it senses your presence.A motorized camera was also seen as impractical because it did not improve Portal’s reliability, Facebook engineers said in a blog post today.Work to build Portal began two years ago as part of the secretive Building 8 project for exploring hardware products at Facebook’s headquarters in Menlo Park, California, Portal team lead and Facebook VP Rafa Camargo told VentureBeat ahead of the device’s debut last fall....

By applying AI and machine learning to multiple data sources—genetic data, EHRs, sensor data, lifestyle data—researchers are taking first steps.

We rarely think too much about noises as we’re listening to them, but there’s an enormous amount of complexity involved in isolating audio from places like crowded city squares and busy department stores. In the lower levels of our auditory pathways, we segregate individual sources from backgrounds, localize them in space, and detect their motion patterns — all before we work out their context.Inspired by this neurophysiology, a team of researchers shared in a preprint paper on Arxiv.org (“Enhanced Robot Speech Recognition Using Biomimetic Binaural Sound Source Localization“) a design devised to test the influence of physiognomy — that is, facial features — on the components of sound recognition, like sound source localization (SSL) and automatic speech recognition (ASR)....

By Lance Eliot, the AI Trends Insider Here’s an age-old question for you, do the ends justify the means? Some trace the origins of this thorny question to the Latin collection Heroides as reportedly written by Ovid (Publius Ovidius Naso), namely it states “Exitus acta probat,” which loosely translated could be interpreted as to whether …

The current generation of AI self-driving cars does not have a certification process akin to standards in the aviation industry,

McAfee and really the whole industry has embraced AI as a key tool to help develop a new set of cyberdefense technologies

If the question is whether the tech industry doing enough to address AI algorithmic biases, the straightforward response is no.

OpenAI has decided not to publish its latest research fearing its potential misuses and the negative societal impact that would have. The institute, backed by the likes of Elon Musk and Peter Thiel, developed an AI which can produce convincing ‘fake news’ articles. Articles produced by the AI writer can be on any subject and […] The post OpenAI withholds its latest research fearing societal impact appeared first on AI News.

As we reported today, Nvidia posted earnings that beat Wall Street’s reduced expectations for fourth fiscal quarter earnings for the period ended January 27. And while that quarter was disappointing, the maker of graphics and AI chips predicted that the year ahead would be a good one, with demand for laptop gaming gear remaining strong.Nvidia CEO Jensen Huang said in an interview with GamesBeat today that revenues came in at lower-than-expected levels in part because Nvidia only had high-end RTX graphics cards available during the quarter. Because of a high-level of inventory from unsold mid-range graphics cards, Nvidia could not launch its new mid-range RTX graphics cards during the quarter....

Nvidia posted earnings that beat Wall Street’s reduced expectations for fourth fiscal quarter earnings for the period ended January 27.Last month, the maker of graphics and AI chips preannounced that the combination of a slow launch for its new RTX graphics cards and the aftermath of the cryptocurrency bust would hurt its earnings. And today, the company reported financial results that were in line with the reduced expectations for the fourth fiscal quarter ended January 31.Nvidia said that non-GAAP earnings per share were 80 cents, compared to the 75 cents expected by analysts. Revenue was $2.21 billion, down 24 percent from a year ago but still above the $2.20 billion that analysts expected. Before Nvidia preannounced on January 28, analysts had expected $2.7 billion....

It’s Valentine’s Day — the holiday on which couples profess their love for one another by gifting chocolates and roses, sharing candlelit meals, and basking together in the glow of a fire (or television). If you’re hitched and somehow procrastinated until now, not to worry — fortunately, artificial intelligence (AI) can help. While robots lack the capacity for affection, they’re more than capable of reserving a floral bouquet, composing a heartfelt melody, and serving up the perfect recipe for a romantic home-cooked meal.That’s all to say if you’re in need of a little machine learning assistance this V-Day, you’re in luck — lots of companies have it on tap. Here’s a roundup of the best we’ve seen....

Deep learning neural networks learn a mapping function from inputs to outputs. This is achieved by updating the weights of the network in response to the errors the model makes on the training dataset. Updates are made to continually reduce this error until either a good enough model is found or the learning process gets …

Tesla is rolling out a 360-degree surveillance system to some of its cars from today, as the company doubles down on its efforts to protect its premium electric vehicles (EVs) from bumps and break-ins.Dubbed “sentry mode,” the new feature will use all eight on-board cameras to provide a holistic view of everything going on outside. This upgrade comes a few months after Tesla introduced a new dashcam feature with version 9.0 of its software; however, that only captured footage from the front-facing camera — and only when the car was active. With sentry mode, not only are all cameras used, but they are functional when the car is inactive (i.e., parked) and they can initiate a number of security-focused deterrents....

Shortly after President Trump issued his vague AI executive order, the US Defense Department outlined a more comprehensive strategy. “The impact of artificial intelligence will extend across the entire department, spanning from operations and training to recruiting and healthcare,” DoD CIO Dana Deasy said. A 17-page document outlines how the DoD intends to advance its […] The post US defense department outlines its AI strategy appeared first on AI News.

Blood tests are generally unpleasant — not to mention costly. On average, getting blood work done at a lab costs uninsured patients between $100 and $1,500. And in the developing world, where the requisite equipment isn’t always readily available, ancillary costs threaten to drive the price substantially higher.That’s why Yossi Pollak, previously at Intel subsidiary Mobileye, and Daniel Levner, a former scientist at Harvard’s Wyss Institute for Biologically Inspired Engineering, founded Sight Diagnostics in 2011. They claim its desktop Olo system can perform “lab-grade,” point-of-care complete blood count (CBC) tests within ten minutes with no more than a pinprick of blood. It’s already available for purchase in much of the European Union, where Sight has obtained the necessary health and safety certification. And now, the Tel Aviv startup is raising money to deploy it globally....

Language is power, and engineering a system that can comprehend it as well as any human is a grand challenge in AI research. Recent contributions like Google’s BERT, a framework that can train state-of-the-art natural language processing (NLP) models in a few hours on a single graphics card, and Facebook’s PyText, which produces over a billion daily predictions for the social network’s apps and services, have nudged the needle forward. But robots capable of speaking naturally, unaided by handcrafted grammar rules and carefully labeled datasets, remain elusive.That hasn’t discouraged OpenAI, an AI research organization backed by tech luminaries Reid Hoffman and Peter Thiel. Over the course of its roughly four-year history, the San Francisco-based nonprofit has investigated autonomous systems that can achieve superhuman performance in Pong and Montezuma’s Revenge and defeat professional Dota players — not to mention imbuing mechanical hands with humanlike dexterity. OpenAI has also published its fair share of work in NLP, and today it is previewing a collection of AI models that can not only generate coherent text given words or sentences, but achieve state-of-the-art (or near-state-of-the-art) performance on a range of NLP tests....

Artificial intelligence (AI) is no longer the stuff of science fiction. Sure, there is still much hyperbole around certain applications of AI (and whether they truly can be called AI), but with smart speakers topping Christmas shopping lists and commercial driverless taxi services now in operation, AI is more science fact than ever.Against that backdrop, a new venture capital (VC) firm officially launched today with $55 million in its coffers for very early-stage startups in the field of AI, machine learning, automation, robotics, and internet of things (IoT).Catapult Ventures is the brainchild of Rouz Jazayeri, a former Intel executive and more recently a partner at Kleiner Perkins; and Darren Liccardo, a former engineering executive at Tesla and drone company DJI. Jazayeri actually helped DJI, one of Kleiner Perkins’ portfolio companies, recruit Liccardo from Tesla, and from this the two began working together on a number of projects....

Here Technologies, the company behind the popular mapping and location platform, has revealed that it’s investing €25 million ($28 million) in a new machine learning research institute that will crunch geolocation data on an industrial scale.The Institute for Advanced Research in Artificial Intelligence (IARAI), as it will be known, will be based in Vienna, Austria. Here stresses that despite its financial backing, the institute will be an independent entity, led by Dr. Sepp Hochreiter, a professor at Johannes Kepler University’s Institute for Machine Learning; and Dr. David Kreil, a data science lead at Vienna’s Boku University.Here will, of course, provide its location-based datasets and wider expertise, while in return it will be able to access any new technologies that emerge from the hub. The IARAI will serve to research machine learning technologies that will “deepen the location intelligence needed to build an autonomous world,” according to a statement issued by the company....

For those without the technical know-how, software development — which more often than not involves a tangled web of commits, reviews, bug testing, and validation — isn’t exactly easy to keep an eye on. Emails and pivot tables help to a degree, but when it comes to deciphering the meandering app design process, the C-suite often finds itself at the mercy of dev team leaders.Appcelerator veterans Jeff Haynie and Nolan Wright hope to change that. They’re the cofounders of Pinpoint, a nearly three-year-old Austin, Texas company developing a data analytics platform for software engineering. Through the use of instrumentation that plays nicely with systems of record like Jira, GitHub, GitLab, BitBucket, and SonarQube, it’s able to translate activity and contributions into actionable, natural language insights that don’t require a computer science degree to understand....

Human eyes are insensitive to polarized light and ultraviolet radiation, but that is not the case for ants, who use it to locate themselves in space. Cataglyphis desert ants in particular can cover several hundreds of meters in direct sunlight in the desert to find food, then return in a straight line to the nest, without getting lost. They cannot use pheromones: they come out when the temperature would burn the slightest drop. Their extraordinary navigation talent relies on two pieces of information: the heading measured using a sort of "celestial compass" to orient themselves using the sky's polarized light, and the distance covered, measured by simply counting steps and incorporating the rate of movement relative to the sun measured optically by their eyes. Distance and heading are the two fundamental pieces of information that, once combined, allow them to return smoothly to the nest....

The answer comes from the study entitled "On the choice of grasp type and location when handing over an object," published in Science Robotics by a research team of The BioRobotics Institute of Scuola Superiore Sant'Anna and the Australian Centre for Robotic Vision. The study reveals the guiding principles that regulate the choice of grasp type during an exchange of objects, encouraging cooperation between a robotic system and a person.The study, conducted in 2018, analysed the behavior of people when they have to grasp an object and when, instead of using it themselves, they need to hand it over to a partner. The researchers investigated the grasp choice and hand placement on those objects during a handover when subsequent tasks are performed by the receiver. Passers tend to grasp the purposive part of the objects and leave "handles" unobstructed to the receivers. Intuitively, this choice allows receivers to comfortably perform subsequent tasks with the objects....

Google says it’s stepping up its efforts to prevent malicious apps from reaching Android smartphones, tablets, and other devices. In a blog post today, Andrew Ahn, product manager at Google Play, said the Mountain View company would continue to improve the automated systems that help root out unscrupulous developers in the Google Play Store, and that it would enhance the backend tools its human moderators use to identify harmful app contents and behaviors.“Google Play is committed to providing a secure and safe platform for billions of Android users on their journey discovering and experiencing the apps they love and enjoy,” Ahn wrote. “To deliver against this commitment, we worked last year to improve our abuse detection technologies and systems, and significantly increased our team of product managers, engineers, policy experts, and operations leaders to fight against bad actors.”...

Relationships are a tricky thing — and particularly fraught in the enterprise world, where the right connection or second-degree contact can make the difference in a high-stakes deal. That’s why almost five years ago, Stanford grads Ray Zhou and Shubham Goel cofounded Affinity, a San Francisco startup developing what they describe as a “relationship intelligence” platform. Unlike some relationship management products on the market, Affinity is entirely automatic — it dispenses with manual data entry in favor of an AI-driven, automated approach. Toward that end, Goel and Zhou say it’s helped to identify over three million “warm introductions” in networks and auto-populated over 300 million data fields with relevant contact information. And it’s just getting started....

Slack today made its Block Kit tool for building more visually appealing apps generally available. The Slack App Directory currently has more than 1,500 apps, and the Slack Platform for the creation of apps and automated bots dates back to 2015.The company also introduced Block Kit Builder for designing apps and testing app prototypes.Block Kit and Block Kit Builder were created to make it easier to design apps and deliver a more consistent user experience. With that in mind, Block Kit launches with five basic blocks, including Image for image containers and Actions to add interactive elements, like buttons and six types of drop-down menus.There’s also a text container block called `Section`, a ‘Context’ container for descriptive metadata, and ‘Divider’ to make space between blocks. ...

China is moving forward in the global “race to 5G,” as state-owned carrier China Mobile has announced (via Xinhua) that it’s already building the first 5G smart highway — a city-scale system of roads capable of supporting cellular network-coordinated transportation services. The infrastructure is currently under construction in Wuhan, the capital of China’s centrally located Hubei Province.As the country’s largest telecommunications company, China Mobile plans to roll out a collection of 5G services on the highway, beginning with “smart toll stations” that could do away with current toll transponders and human operators. The carrier also plans to gather real-time traffic information and make AI-assisted predictions using the data, as well as supporting autonomous cars....

As machine learning expands into climate modeling, EAPS Associate Professor Paul O’Gorman answers what that looks like and why it's important now.

Amazon is making voice apps created by Echo owners available in the Alexa Skills Store. Like the more than 80,000 Alexa skills available now, Skill Blueprints can be rated and reviewed by the larger user community. In contrast to the more technical process through which the average Alexa skill is made, Skill Blueprints are made with easy-to-use templates that require no coding, allowing anyone to make their own voice app.Blueprints in the Alexa Skills Store will first be introduced for users in the United States.New Blueprints made especially for bloggers, content creators, and other organizations were also introduced today. These can be used by personal trainers and fitness studios, for example, or by anyone interested in reaching out to clients or customers. Storyteller Blueprints were also introduced today and run the gamut from fairy tales and spooky stories to adventure, sci-fi, and stories about a family member....

Personalized marketing speaks to people. It really does. According to a recent Adobe survey, customers were 51 percent more likely to make a purchase and 49 percent more likely to become loyal to a company if they could identify with the advertising in some way. The trouble is, crafting bespoke campaigns takes a lot of time and money that most agencies and brands don’t have.That’s where the new and improved Adobe Experience Manager (a part of Adobe Marketing Cloud in Adobe Experience Cloud) comes in — or so goes the sales pitch. Thanks to Sensei, Adobe’s cross-platform machine learning technology, Adobe says the newest version of Experience Manager — its content management solution for building websites, mobile apps, and forms — can quickly serve up videos, photos, and other assets with no more than a tag or similar media clip. And that’s just the tip of the iceberg....

Edge computing company Xnor.ai today announced that it has made a device capable of deploying a state-of-the-art deep learning algorithm with a small solar panel — without requiring a battery or other power source.Equipped with a camera mounted to a field programmable gate array (FPGA) board, the always-on device uses edge computing, which means photos or video data captured by the camera never leaves the device.To alert a user when the algorithm detects or classifies a particular object, a low data rate wireless communication protocol can be used to send a text message.The device introduced today can also be attached to a battery or other source of power, but a solar panel was used to demonstrate unique use cases that can emerge when an AI system produces reliable results and consumes 5mJ of energy for every inference instance....

The dozens of companies vying for a foothold in the $11.2 billion drone industry — and by extension the $400 million same-day delivery market — have all made more noise than Elroy Air, a San Francisco startup cofounded almost three years ago by former Airwave product manager Clint Cope and Sifteo cofounder David Merrill. Elroy — the namesake of Elroy Jetson in the Jetsons — operated largely under the radar until December 2017, when it emerged from stealth with a prototype aircraft that could carry hundreds of pounds of cargo over hundreds of miles.It has kept its head down since then, but it’s reemerging now with news that it has closed a $9.2 million seed round of funding from new and existing investors Catapult Ventures, Precursor Ventures, Lemnos, Haystack, Shasta Ventures, Levitate Capital, 122West, Amplify Partners, and Homebrew. Elroy also took the wraps off of its latest creation — the unmanned vertical-takeoff-and-landing (VTOL) drone Elroy Air Chaparral — and revealed that it plans to kick off real-world testing of its drone fleet this year, preceding the launch of commercial programs with a small number of shippers in select locations in 2020....

President Donald Trump has issued his executive order relating to AI, only it misses something quite important to ensure US leadership. The executive order titled ‘Maintaining American Leadership in Artificial Intelligence’ claims to be guided by five principles: Driving technological breakthroughs. Leading in developing standards while reducing barriers preventing safe AI testing and deployment. Training […] The post Trump’s AI executive order misses something important appeared first on AI News.

The amount of data generated today boggles the mind — U.S. companies alone produce 2.5 quintillion bytes daily, enough to fill ten thousand Libraries of Congress in a year — and much of it is of the time-series variety (i.e., data points indexed in time order). Given the sheer volume, it’s no wonder that only 12 percent of companies say they’re analyzing the data they have, according to Forrester Research.That’s one of the reasons Paul Dix — who’s helped to build software for startups, large companies, and organizations like Microsoft, Google, McAfee, Thomson Reuters, and Air Force Space Command — founded Y Combinator- and Bloomberg Beta-backed InfluxData (formerly Errplane) in 2012. The San Francisco startup develops an open source time series platform, InfluxDB, that is optimized to handle metrics and events in DevOps, internet of things (IoT), and real-time analytics domains. And after a banner year that saw revenue double, InfluxDB 2.0 launch in alpha, and Flux — a functional language for both querying and processing data — debut in technical preview, the startup is gearing up for growth....

How can the average restaurant owner hope to make sense of the capricious culinary industry when restauranteurs, culinary directors, and multinational corporations alike have a tough go of it? There are billions of dollars at stake (over $799 billion with respect to restaurants) in a market with razor-thin margins; a typical sit-down joint takes home just 3 to 5 percent in profit. It’s a lot of pressure.Alon Chen, formerly Google’s chief marketing officer for Israel and Greece and global lead for the World Economic Forum, set out nearly two years ago to bring predictability to the frequently tumultuous world of food and beverage. Tastewise, which formally launches today, is the fruit of his and cofounder Eyal Gaon’s labor: a Tel Aviv-based startup with a platform that combines artificial intelligence (AI), predictive analytics, computer vision, and natural language processing to uncover gastronomic insights for restaurants, hospitality groups, and brands....

There’s a healthy demand for driverless trucks. It’s an industry predicted to reach 6,700 units globally, totaling $54.23 billion this year, and one that stands to save the logistics and shipping industry $70 billion annually while boosting productivity by 30 percent. Besides cost savings, the growth is driven in part by a shortage of human drivers. In 2018, the American Trucking Associates estimated that 50,000 more truckers were needed to close the gap in the U.S., even despite the sidelining of proposed U.S. Transportation Department screenings for sleep apnea.TuSimple is one of the companies vying for a slice of the market. For the better part of four years, the China-based startup has been developing trucks at its R&D lab in San Diego and test operations facility in Tuscon that can drive depot-to-depot without human intervention. By all accounts, it’s made remarkable progress: TuSimple is currently taking three to five fully autonomous, revenue-generating trips a day for a dozen customers (including Fortune 100 companies and “household names”) on three (and soon four) different routes in Arizona, and will by June expand the size of its U.S. fleet from 11 to 50 trucks. It hopes to end 2019 with a 200-truck fleet in the U.S. and 300-truck fleet in China, which would make it the largest self-driving truck solutions company in the world....

Seattle, Washington data analytics company Tableau wants to make it easier to get straightforward answers to sales questions. Toward that end, it today announced the general availability of Ask Data, a natural language processing feature for Tableau Server and Tableau Online that enables users to pose queries and get a response. It also debuted Tableau Data Management — a new subscription add-on targeting data prep scenarios — and a pair of revamped smartphone clients.Ask Data is available as a part of Tableau‘s latest release, Tableau 2019.1, and supports the platform’s full range of live and extract data sources.Using Ask Data to get an answer is as simple as typing in a question like “What were my sales this month?”, after which Tableau will return an interactive visualization that can be explored with clarifying follow-ups, new questions, and drag-and-drop gestures....

The U.S. last-mile transportation market is more lucrative than you might think — some estimates peg it at $28 billion. Multimodal companies like Lime and Bird have raised collective hundreds of millions of venture capital for e-scooters, bikes, and compact electric vehicles well-suited for short trips in high-density urban cores, and yet another company that’s experienced outsized growth is May Mobility. The Ann Arbor, Michigan-based company was cofounded two years ago by Alisyn Malek, former Ford autonomous driving head; Toyota Research Institute codirector Edwin Olson; and Steve Vozar, who previously oversaw the University of Michigan’s robotics lab. Its six-seat electric shuttles have been deployed in cities across the Midwest, and with a new round of financing, it’s gearing up for nationwide expansion....

Mozilla and Ubisoft today announced a partnership to develop Clever-Commit, an AI coding assistant. Clever-Commit is designed to help programmers evaluate if a code-change will introduce a new bug, based on learnings from past bugs and fixes. The tool combines data from the bug tracking system and the version control system to detect patterns of programming mistakes based on the history of the development of a given piece of software. Yves Jacquier, head of Ubisoft La Forge, unveiled the partnership at Dice Summit 2019 in Las Vegas.If Clever-Commit sounds familiar, you might be remembering its predecessor, a prototype named Commit-Assistant. Ubisoft La Forge, Ubisoft Montreal’s research lab, and Concordia University, unveiled Commit-Assistant to the world in March 2018. After successful tests with data accumulated during the development of video games, the prototype was transferred to the Ubisoft Technology Group with the goal of being used across the company. ...

Data collected by vehicles through cameras, LiDAR and GPS allow the researchers to capture video snippets of humans in motion and then recreate them in 3D computer simulation. With that, they've created a "biomechanically inspired recurrent neural network" that catalogs human movements.With it, they can predict poses and future locations for one or several pedestrians up to about 50 yards from the vehicle. That's at about the scale of a city intersection."Prior work in this area has typically only looked at still images. It wasn't really concerned with how people move in three dimensions," said Ram Vasudevan, U-M assistant professor of mechanical engineering. "But if these vehicles are going to operate and interact in the real world, we need to make sure our predictions of where a pedestrian is going doesn't coincide with where the vehicle is going next."...

The capacity of a deep learning neural network model controls the scope of the types of mapping functions that it is able to learn. A model with too little capacity cannot learn the training dataset meaning it will underfit, whereas a model with too much capacity may memorize the training dataset, meaning it will overfit …

Deep reinforcement learning — an algorithmic training technique that drives agents to achieve goals through the use of rewards — has shown great promise in the vision-based navigation domain. Researchers at the University of Colorado recently demonstrated a system that helps robots figure out the direction of hiking trails from camera footage, and scientists at ETH Zurich described in a January paper a machine learning framework that aids four-legged robots in getting up from the ground when they trip and fall.But might such AI perform just as proficiently when applied to a drone rather than machines planted firmly on the ground? A team at the University of California at Berkeley set out to find out....

IBM’s Watson artificial intelligence systems can now be used on Google Cloud Platform, AWS, Microsoft’s Azure, and other cloud providers through an integration with IBM Cloud Private for Data. The approach can also be used to bring Watson to private clouds or on-premise servers.Services like Watson Studio for building AI, Watson Assistant, and AI OpenScale were previously only available through IBM’s public cloud.The move was made, IBM Data and AI Group general manager Rob Thomas told VentureBeat in a phone interview, to help organizations save money and time and apply AI wherever they store their data.“What’s unique about what we’re doing — and I’m not sure if people will follow suit because I don’t think it’s their strategy — we are saying we are going to use open-source Kubernetes, enable portability,” Thomas said. “And for most of the public cloud providers, portability is not in their interest, so I think we’ll probably be unique in that stance.”...

We’ll explore each of these five dimensions—accuracy, fairness, explainability, stability, and adoption—as they apply to AI in the public sector.

We might decide to build a robot that can get into and out of the car, akin to a human being, and the robot will be imbued with the ability to drive a car.

AI’s ability to diagnose melanoma with higher fidelity illustrates the potential impact on patient outcomes; still there are challenges.

The core idea underlying transfer learning is that researchers can “recycle” the knowledge gained from a large, pre-trained neural network for a new task.

Datadog, the heavily-funded cloud monitoring platform for applications and infrastructure, has acquired under-the-radar French startup Madumbo, which develops an AI-powered web app-testing service. Terms of the deal were not disclosed.Founded in 2010, New York-based Datadog’s DevOps toolset enables developers to monitor everything in their stack, aggregating metrics and events from across all their databases, servers, apps, and more under a single unified view. The company claims some big name clients, such as Sonos, WeWork, Intel, Ubisoft, Samsung, Zendesk, Stripe, eBay, and Target.Founded out of Paris in 2017, Madumbo’s automated platform helps companies determine whether a web app is performing as it should do. Its bot checks for errors using code run from a real browser, meaning that it should detect the same issues that end users experience....

IBM’s argumentative AI-powered computer has faced off in a debate against a human professional and lost. Computers typically have a vast amount of information at their disposal but IBM’s was unable to persuade a live audience of its viewpoint. IBM’s debating computer was tasked with taking on Harish Natarajan, a grand finalist in 2016’s World […] The post Human beats IBM’s AI-powered debate computer appeared first on AI News.

Import.io, a Los Gatos, California-based startup developing a toolset that can automatically scrape and process web data, today announced that it’s acquiring enterprise data extraction company Connotate for an undisclosed sum. The deal, Import.io CEO Gary Read said, will enable the company to scale and meet “fast-growing” enterprise demand, bolster its IP (Connotate has eight patents to its name), and allow it to “further solidify” its leadership in a market that’s forecast to be worth $5 billion in 2019.“With Connotate, Import.io is acquiring blue-chip enterprise customers, innovative technology, and outstanding domain expertise, further establishing us as the undisputed leader in web data integration,” Read added. “External spend on web data extraction is growing 70 percent annually as organizations shift from internal engineering-led projects to enterprise-class SaaS. This acquisition will help us to continue to rapidly scale to meet the exploding demand for our solution and, most importantly, will further enable us to help our customers achieve success and competitive advantage. All customers, whether currently with Connotate or Import.io, will benefit.”...

Nearly three years ago, Chetan Conikee, Manish Gupta, and Vlad A. Ionescu put their collective heads together to develop a software platform that provided robust, code-level security to app developers without adversely impacting workflows. It evolved into software called ShiftLeft (and a namesake Santa Clara startup), which combines vulnerability-detecting static code analysis with app-preserving instrumentation. Since ShiftLeft’s emergence from stealth in 2017, the company has expanded to meet what it describes “healthy” demand for its product, and today, it announced new funding that’ll lay the groundwork for future growth.ShiftLeft says that it has secured $20 million in series B financing led by Thomvest Ventures and new investor SineWave Ventures, as well as existing investors Bain Capital Ventures and Mayfield. It comes less than 18 months after the company raised $9.3 million in a series A round, and brings its total capital raised to nearly $30 million....

Mere hours after the beta launch of LinkedIn Live, an invite-only video broadcast service for conferences and events, LinkedIn took the wraps off a new product geared toward recruiters and job seekers. This morning, the Microsoft subsidiary previewed the Intelligent Hiring Experience, a dashboard that collates LinkedIn’s existing hiring suite — comprising LinkedIn Recruiter, LinkedIn Jobs, and Pipeline Builder — in a streamlined, built-from-the-ground-up interface.The new feature is scheduled to launch in late summer 2019 with over 15 product enhancements, several of which will roll out in the coming months.“[There are] simply more people on LinkedIn looking for jobs [than] there ever have been,” John Jersin, vice president of Talent Solutions and Careers at LinkedIn, told reporters during a dial-in conference on Monday. “This new platform will allow all of [our] products to learn from each other so that matches in each system get more intelligent as you use our various products. And the new interface that we built on this platform will make it seamless to move from one step in the hiring process and one product straightaway to another.”...

You can tell a lot about a person from their facial features and expression — whether they’re displeased or frustrated, for example, and their relative age and weight. That’s why popular smartphones employ facial recognition as their authentication method of choice, and why Palo Alto-based identity-as-a-service provider Jumio’s new product offering — Jumio Authentication — uses facial data to verify users’ identities.Jumio Authentication, which formally launches today, is a tool for businesses to prove users are who they say they are. It combines biometrics for identity proofing and “ongoing” 3D face authentication, courtesy Nevada-based biometric company FaceTec’s Zoom 3D Face Login technology. Users first snap a photo of their driver’s license, passport, or ID card, and then use a mobile device camera or webcam to capture their faces. In the course of authentication, Jumio Authentication compares the “selfie” — a 3D face map containing 100 times the liveness data of a 2D photo — to the picture on the aforementioned ID, and retakes the selfie for good measure,...

A quick scan of the venture capital (VC) landscape reveals many funds have a specific raison d’être that helps summarize it in a nutshell: some are focused on one specific region, for example, while others are geared toward a specific kind of technology or stage-specific company.London-based InReach Ventures’ big pitch is that it is the “AI-powered VC firm,” and today it announced the closing of its first proper fund at €53 million ($60 million).One thing most VC firms have in common is their basic investment methodology — they employ research and due diligence to evaluate the risks of an investment opportunity. These risks vary depending on what stage a company is at, of course, with seed or series A-stage ventures generally more risky than profit-making growth-stage companies that are already mulling an exit. But there is one challenge that is orders of magnitude greater than due diligence: how do VCs actually find companies worth investing in to begin with?...

IBM’s Project Debater, an AI system that can engage in live debate with humans, today faced off with 2016 World Debating Championships Grand Finalist and 2012 European Debate Champion Harish Natarajan. The debate was about preschool subsidies, with Project Debater arguing in favor and Natarajan arguing in the opposition.More than 50 percent of the audience at the Yerba Buena Center for the Arts in San Francisco said in a poll taken after the debate that Debater enriched their knowledge on the subject, while about 20 percent said the same of Natarajan’s argument.The two appeared to fight to a draw when it came to changing minds in the audience: Following the debate, 17 percent of the audience against the motion changed their minds, and 17 percent of the audience in favor of preschool subsidies changed their mind....

Andela, a company that connects developers in Africa with growing U.S. startups like SeatGeek and Headspace, will incorporate artificial intelligence into its recommendation engine to create better matches between talent and employers.“We’re working on the algorithms now and I expect it to be an active part of how we operate by the end of this year,” Andela CEO Jeremy Johnson told VentureBeat in a phone interview.Last month, the company raised $100 million in a round led by Al Gore’s Generation Investment Management, one of the largest venture-funded rounds ever for a startup operating in Africa.Bringing machine intelligence to its recommendation engine and making new tools to help managers evaluate the performance of remote engineering teams and employees will be a main focus of the funding, Johnson said....

New system of “strain engineering” can change a material’s optical, electrical, and thermal properties.

MIT associate professor of aeronautics and astronautics describes the seamless flow of people, things, and materials.

"Many researchers need to identify and collect data on specific individuals in their work, for example to estimate survival, reproduction, and movement," said Derek Lee, associate research professor of biology at Penn State and principal scientist of the Wild Nature Institute. "Instead of tags and other human-applied markings that could interfere with the animal's behavior, many researchers take photographs of the animal's unique markings. We have pattern recognition software to help analyze these photos, but the photos all have to be manually prepared for analysis. Because we often have thousands of photos to go thorough, this creates a serious bottleneck."Lee uses photographs as part of a large ongoing study to understand births, deaths, and movement of more than 3,000 giraffes in East Africa. He and his team take digital photographs of each animal's unique and unchanging spot patterns to identify them throughout their lives. But before the images can be processed by pattern recognition software to identify individuals, the research team has to manually crop each photo or delineate an area of interest. Lee collaborated with scientists from Microsoft, who have provided a new image processing service to automate this essential and time-consuming process using machine learning technology deployed on the Microsoft Azure cloud....

How do humans learn to cooperate? It’s an interesting question, one behavioral anthropologists have been studying for decades. Social norms — that is, common understandings or informal rules, like dining etiquette and fashion sense — are thought to play a part, but it’s tough to measure the extent to which they shape society and how they’re affected by other factors.Fortunately, that’s where artificial intelligence (AI) comes in.In a newly published paper on the preprint server Arxiv.org (“Understanding The Impact of Partner Choice on Cooperation and Social Norms by means of Multi-agent Reinforcement Learning“), scientists describe an AI system trained using reinforcement learning — a technique that uses rewards to drive agents toward goals — for understanding how an interactions within a society affect the overall societal outcome....

As AI expands access to clean, rich data, governance, privacy, and protocol issues become more urgent, especially around privacy, bias, and fairness. To learn more about the kind of actionable insights AI delivers and how to avoid the pitfalls when launching AI strategies, don’t miss this VB Live event!Register here for free.Big data and AI is a booming business — with over 97 percent of companies interested in investing in AI initiatives, the compound annual growth rate of spending on big data and analytics has climbed to 11.9 percent, and revenues will rack up to more than $210 billion by 2020.But while 78 percent of companies are pretty confident about their data maturity, confidently rating themselves as “medium” or “high” when it comes to their in-house skills, tech, and initiatives, only 12 percent actually qualify as mature. The advantage of AI-powered enterprise analytics are huge, from increased agility and better decision-making to reduced costs after the initial investment as well as improved fraud detection....

President Trump is expected to sign an executive order on Monday meant to spur the development and regulation of artificial intelligence, technology that many experts believe will define the future of everything from consumer products to health care to warfare. A.I. experts across industry, academia and government have long called on the Trump administration to …

Researchers from Google, Microsoft, and IBM have joined together to create Responsible AI Licenses (RAIL). With its first two offerings available in the past two weeks, RAIL now offers end user license agreements and source code license agreements that software providers, researchers, and developers can include with their software to prevent harmful use of their technology.RAIL may create additional license agreements based on input from AI researchers and the broader AI community.“We recognized the risks our work can sometimes bring to the world that led us to think about potential ways of doing this,” said cofounder Danish Contractor, who works at IBM Research in New Delhi during a phone interview with VentureBeat. “Our goal is to empower developers to be able to restrict the use of their AI technology and to prevent harmful and irresponsible use.”...

Predicting customers’ purchasing patterns is easier said than done. It necessarily involves collecting huge amounts of data, and growing and maintaining the resulting datasets over the course of months (or years). And that’s only the start — sussing out the corpora’s key metrics requires purpose-built sophisticated, scalable analytics tools.Former BioWare senior systems architect Mike Babineau and Electronic Arts analytics product manager Lillian Chou were inspired to tackle the problem while working at game developer and publisher Rumble Entertainment, where they engineered large-scale infrastructure and pipelines to query data. Their work with complex datasets laid the groundwork for Second Measure, a San Mateo-based Y Combinator alumnus developing a self-service “data exploration” platform that supports both daily and real-time transactional tracking....

Nuro, a startup targeting the multibillion-dollar autonomous delivery robot market, has attracted sizeable venture capital from Softbank. The Japanese holding conglomerate today announced that it has invested almost $1 billion — $940 million — in Nuro through its Vision Fund, at a valuation of around $2.7 billion. That’s up substantially from the $92 million in series A financing Nuro secured last year.Nuro was cofounded in 2016 by Dave Ferguson and Jiajun Zhu, both veterans of the secretive Google self-driving car project that eventually spun out as Waymo. Currently, the Mountain View, California-based company has about 200 employees and 100 contract workers and is working on a pilot program with Kroger to ferry goods between shoppers and a few of the grocery giant’s 2,800 stores in 35 states. (A Kroger-owned Fry’s Foods store in Phoenix began using Nuro’s technology in December.)...

Professor Judith Barry’s attention-grabbing installations create thought-provoking experiences for viewers.

According to several outlets briefed over the weekend on the White House’s plans, President Trump will today sign an executive order establishing a program — the American AI Initiative — that’ll task federal agencies with devoting more resources to artificial intelligence (AI) research, training, and promotion. It comes after President Trump promised  “investments in the cutting-edge industries of the future” during his State of the Union speech last week, and after companies like Amazon, Microsoft, and Google called for guidelines governing the use and development of AI technologies such as facial recognition.“AI is something that touches every aspect of people’s lives,” a White House official told Reuters. “What this initiative attempts to do is to bring all those together under one umbrella and show the promise of this technology for the American people.”...

Want to dive earnestly into artificial intelligence (AI) development, but find the programming piece of it intimidating? Not to worry — Uber has your back. The ride-hailing giant today debuted Ludwig, an open source “toolbox” built on top of Google’s TensorFlow framework that allows users to train and test AI models without having to write code.Uber says Ludwig is the culmination of two year’s worth of work to streamline the deployment of AI systems in applied projects, and it says it’s been tapping the tool suite internally for tasks like information extraction from driver licenses, identification of points of interest during conversations between driver-partners and riders, food delivery time prediction, and more....

In Financial Times op-ed, MIT president says higher education must teach students to be “AI bilingual.”

WiTricity, a venture capital (VC)-backed startup that’s developing wireless charging technology for electric vehicles (EVs), has acquired Qualcomm’s EV wireless charging unit Halo.Terms of the deal were not disclosed, but Qualcomm said it will become a minority shareholder in WiTricity as a result of the transaction.While the EV market is still modest compared to that of petrol and diesel, electric vehicles are expected to accelerate significantly over the next decade — EV sales grew by more than 80 percent last year in the U.S. alone, and major carmakers are already committed to making hybrid or electric cars only from this year.But one of the big challenges facing EV makers is the issue of charging: In short, it can be a real pain. That is why a number of companies have been looking at ways to not only extend the range cars can cover on a single charge, but to make the charging process itself seamless....

IBM Research has a long history of inventing the future. So the big tech company’s researchers take their predictions seriously, and today they are revealing their annual “5 in 5” predictions, which detail five innovations that will change our lives in the next five years.IBM will talk about the predictions at its Think 2019 event in San Francisco on Wednesday from 10 a.m. Pacific time to 11 a.m. Arvind Krishna, senior vice president of IBM Cloud & Cognitive Software, said in a blog post that the world’s population will cross eight billion for the first time.“Our complex food chain—already stressed by climate change and a finite water supply—will only be tested further,” he wrote. “To meet the demands of this crowded future, we will need new technologies and devices,scientific breakthroughs,and entirely new ways of thinking about food safety and security.”...

Edge computing — that is, network architectures in which computation is relegated to smart devices as opposed to servers in the cloud — is forecast to be a $6.72 billion market by 2022. Its growth will coincide with that of the deep learning chipset market, which some analysts predict will reach $66.3 billion by 2025. They’re complementary, as it turns out — edge computing is projected to make up roughly three-quarters of the total global AI chipset business in the next six years.David Schie, a former senior executive at Maxim, Micrel, and Semtech, thinks both markets are ripe for disruption. He — along with WSI, Toshiba, and ARM veterans Robert Barker, Andreas Sibrai, and Cesar Matias — cofounded AIStorm in 2011, a San Jose artificial intelligence (AI) startup that develops chipsets which can directly process data from wearables, handsets, automotive devices, smart speakers, and other internet of things (IoT) devices. It today emerged from stealth with $13.2 million in Series A backing from biometrics supplier Egis Technology, imaging sensor firm TowerJazz, Meyer Corporation, and Linear Dimensions Semiconductor — all four of which say they plan to integrate the company’s technology into upcoming products....

A report from the NHS suggests the impending technological ‘revolution’ in healthcare will increase the amount of time doctors can spend with patients. NHS doctors are overburdened; a problem only getting worse from a growing and ageing population, and not enough funding. The report was led by US academic Eric Topol and calls for a […] The post NHS report suggests AI will give docs more patient time appeared first on AI News.

Central Banks, especially the Federal Reserve Bank, love to reference models to justify their rate decisions. And, while much of the business community and many economists question the relevance of the Philips Curve, the Fed still uses this model. But the model is outdated already and will become more out of sync as new innovations such as AI continue to transform our economy and workforce.From a certain perspective, one can understand the Fed’s decision to continue using the Philips Curve, given their dual mandate of full employment and price stability; the Philips Curve is a model that includes both. However, “the curve” was based on studying unemployment and wage inflation between the years of 1861-1957 in the United Kingdom (which by today’s standards would be considered a relatively closed economy). As a result, the Philips Curve becomes less relevant with each passing year. It does not take into account the impact of globalization (wage arbitrage), internet commerce (the ability to purchase like goods out of market), or the labor force participation rate (impacted by changing demographics and attitudes as well as skill set mismatches). All three of these factors exert an increasing influence on both wages and employment levels in a way that is dynamic and asymmetric over time....

Modern deep learning libraries such as Keras allow you to define and start fitting a wide range of neural network models in minutes with just a few lines of code. Nevertheless, it is still challenging to configure a neural network to get good performance on a new predictive modeling problem. The challenge of getting good …

Machine intelligence doesn’t automatically lead to smarter user experience if product designers and machine learning experts don’t talk the same language.The language and concepts of machine learning are far from intuitive. And user experience design requires an understanding of how people think and behave, simultaneously taking into account the irrationality of human behavior and the messiness of everyday life.Because of the different skills these two disciplines require, it’s normal to see user experience designers and machine learning experts work in their own separate silos even though they’re building the same product. Often, experts from both fields are not familiar with each other’s methods and tools and so are unable to grasp what can be achieved by combining experience design with machine learning. To break these professional silos, the product team needs to make a steadfast and conscious effort, but how to get started?...

The rise of voice assistants has put a new customer-facing channel on the map for marketers. The promise is clear: voice-enabled devices operate as a fine blend of digital and physical realities, which makes room for truly contextual interactions with users. But with voice technology so young, it’s still unclear if it can add up meaningfully to the marketing toolkit.In 2016, Gartner predicted 75 percent penetration of voice-activated devices in US households by 2020. We’re now two years short of this deadline and the penetration rate is a humble 13 percent (though expected to rise to 55 percent by 2022).2018 is the year voice-enabled devices hit a one-billion-sold mark. And it’s not only about smart speakers per se; this adoption wave is spilling over to voice assistants installed on smartphones, smart watches, and IoT-connected devices....

It’s mind-blowing how much world-shaping work that gets done in hotel ballrooms. Machine learning experts regularly gather at conferences around the world to discuss noteworthy work and how to move the industry forward. Few are fortunate enough to attend in person, but you can sometimes find video online.The most recent example: The Association for the Advancement of Artificial Intelligence (AAAI) met in Hawaii last week, and among topics discussed was the roadmap for AI research in the United States for the next 20 years.The process to create a plan for the next two decades started in November with private workshops attended by academics and people from industry. A draft of the report is due out later this month, and the final will come in April following a period of public comment....

Cyber threats regularly overwhelm traditional security solutions. It’s growing clear that artificial intelligence and machine learning is the safest path to lock down data and protect the enterprise. Learn how new AI-powered security can protect companies and their customers better when you join this VB Live event! Register here for free.The amount of information that we have to pour through in order to identify threats and vulnerabilities and ongoing attacks is growing non-linearly, says Fernando Maymi, Ph.D., CISSP, a security practitioner with over 25 years’ experience in the field for both government and private sector organizations in the US and abroad.“What AI fundamentally does is give us a fighting chance,” Maymi says....

Facebook has acquired visual search and AI startup GrokStyle, a company that previously worked with retailers to make computer vision that helps people shop.“Our team and technology will live on, and we will continue using our AI to build great visual search experiences for retail,” the company said in a post on its website.As part of the acquisition, GrokStyle will wind down its existing business.“We are excited to welcome GrokStyle to Facebook. Their team and technology will contribute to our AI capabilities,” Facebook spokesperson Vanessa Chan said in a statement shared with VentureBeat.GrokStyle technology could be used in a number of ways at Facebook, from its Marketplace where users buy and sell goods to computer vision systems used for moderation of content on Facebook or Instagram....

“The purpose of a business should be to solve real problems for real customers,” says MIT Professor Hari Balakrishnan, cofounder of Cambridge Mobile Telematics.

A team led by Sari Nijssen of Radboud University in Nijmegen in the Netherlands and Markus Paulus, Professor of Developmental Psychology at Ludwig-Maximilians-Universitaet (LMU) in Munich, have carried out a study to determine the degree to which people show concern for robots and behave towards them in accordance with moral principles. Their findings appear in the journal Social Cognition.According to Sari Nijssen, the study set out to answer the following question: "Under what circumstances and to what extent would adults be willing to sacrifice robots to save human lives?" The participants were faced with a hypothetical moral dilemma: Would they be prepared to put a single individual at risk in order to save a group of injured persons? In the scenarios presented the intended sacrificial victim was either a human, a humanoid robot with an anthropomorphic physiognomy that had been humanized to various degrees or a robot that was clearly recognizable as a machine....

While the rise of AI presents an opportunity for executives in every industry, implementing a company-wide AI strategy is challenging.

Amazon has put its weight behind the growing number of calls from companies, individuals, and rights groups to establish facial recognition standards. Michael Punke, VP of Global Public Policy at Amazon Web Services, said. “Over the past several months, we’ve talked to customers, researchers, academics, policymakers, and others to understand how to best balance the […] The post Amazon joins calls to establish facial recognition standards appeared first on AI News.

An initial influx of AI rockstar developers from numerous university-related AI self-driving car efforts have made their way into the industry.

Singapore has released a framework on how AI can be ethically and responsibly used, a “living document” model framework for AI governance.

Castle, a San Francisco-based startup that wants to help businesses keep their customers’ online accounts safe from fraud, has raised $9.2 million in a series A round of funding from Index Ventures, with participation from Y Combinator, First Round Capital, F-Prime Capital Partners, and a host of individual angel investors.Founded in 2015, Castle works with web and app developers looking to offer greater security inside their consumer-facing apps. Castle’s technology helps prevent all manner of account takeover (ATO) efforts, whether through manual attempts or automated methods, including credential stuffing.In 2017, identify fraud cost 17 million U.S. consumers $17 billion, according to a report by Javelin Strategy & Research, with “account takeover” making up more than $5 billion of the losses. A separate report last year by Shape Security noted that credential stuffing, specifically, costs U.S. businesses around $5 billion a year. While all the big banks and major technology companies such as Apple, Amazon, and Google have resources to protect users’ online accounts from third-party chicanery, the same can’t always be said for smaller companies. And that is precisely the problem Castle is looking to fix....

A San Francisco lawmaker is proposing what would be a nationwide first: a complete moratorium on local government use of facial-recognition technology.

Here’s a look at four current use cases where AI is having a substantial impact right now, including some you’d never have imagined just a few years ago.

Amazon’s Alexa can detect whispered speech — that’s how it knows when to whisper back. But what about AI that’s capable of sussing out frustration? Enter MIT Media Lab spinoff Affectiva’s neural network, SoundNet, which can classify anger from audio data in as little as 1.2 seconds regardless of the speaker’s language — just over the time it takes for humans to perceive anger.Affectiva’s researchers describe it (“Transfer Learning From Sound Representations For Anger Detection in Speech“) in a newly published paper on the preprint server Arxiv.org. It builds on the company’s wide-ranging efforts to establish emotional profiles from both speech and facial data, which this year spawned an AI in-car system codeveloped with Nuance that detects signs of driver fatigue from camera feeds. In December 2017, it launched the Speech API, which uses voice to recognize things like laughing, anger, and other emotions, along with voice volume, tone, speed, and pauses....

Neural networks — layered functions that mimic the behavior of neurons in the brain — are good at lots of things, like predicting floods, estimating heart attack mortality rate, and classifying seizure types. But they hold particular promise in the text-to-speech (TTS) realm, as evidenced by systems like Google’s WaveNet, Baidu’s DeepVoice, and WaveLoop. Another case in point: A new artificially intelligent (AI) TTS system created by researchers at Facebook that can produce speech in multiple languages.The team describes their work in a paper (“Unsupervised Polyglot Text-to-Speech“) published on the preprint server Arxiv.org.“The … network is able to transfer a voice, which was presented as a sample in a source language, into one of several target languages,” they wrote. “We would like to take a sample of a speaker talking in one language and have her voice-avatar speak, as a native speaker in another language.”...

An interesting benefit of deep learning neural networks is that they can be reused on related problems. Transfer learning refers to a technique for predictive modeling on a different but somehow similar problem that can then be reused partly or wholly to accelerate the training and improve the performance of a model on the problem …

Microsoft today announced the general availability of Azure Data Explorer (ADX) and Azure Data Lake Storage Gen2 (ADLS Gen2) — two services it says will afford Azure customers greater flexibility in managing unstructured data, or data generated from interactions on the web, software-as-a-service apps, social media, mobile apps, and internet of things devices. The company also launched a preview of Mapping Data Flow in Azure Data Factory, a visual, no-code app that enables developers to build “data transformations.”“We always strive to make it very easy for IT staff to adopt analytics and for line-of-business people to utilize and deliver powerful insights using beautiful products,” said John Chirapurath, general manager of Azure data, blockchain, and AI at Microsoft....

The US is planning an intervention that will see it play a role in ensuring the country leads in technologies such as AI and 5G. In an email to The Hill, an official wrote: “Within the coming weeks, we could expect to see action designed to preserve American [research and development] leadership in artificial intelligence, […] The post White House will intervene to ensure the US leads in AI, 5G appeared first on AI News.

Amazon supports the creation of a “legislative framework” covering facial recognition technology. That’s according to Michael Punke, vice president of global public policy at Amazon’s AWS division, who penned a blog post this week outlining proposed guidelines for the “responsible use” of face-classifying software by private, commercial, and government entities.“Over the past several months, we’ve talked to customers, researchers, academics, policymakers, and others to understand how to best balance the benefits of facial recognition with the potential risks,” Punke wrote. “It’s critical that any legislation protect civil rights while also allowing for continued innovation and practical application of the technology … We encourage policymakers to consider these guidelines as potential legislation and rules are considered in the U.S. and other countries.”...

Following this week’s leak of the first official image of Samsung’s sequel to the Gear Sport, additional firmware and image leaks appear to have established the sleek wearable’s name, specs, and color variations. The smartwatch will apparently be known as the Galaxy Watch Active and include a considerable number of hardware upgrades over the Gear Sport, albeit with a much smaller battery.Apart from losing the physical turning bezel of its predecessor, Galaxy Watch Active is set to improve upon everything from screen size to sensors and wireless capabilities. According to XDA Developers, the new watch will jump from a 1.2-inch screen to a 1.3-inch display with the same 360 by 360 resolution, add a speaker, and come in an LTE cellular-capable version. It will also include new sensors for linear acceleration, gravity, and stress monitoring, as well as support for faster 1-Amp charging....

A chatbot to complement a doctor’s visit? It’s not as radical as it sounds. Hospitals and health systems have plenty of incentive to further the adoption of AI conversational agents; according to a survey published by research firm Statista, 27 percent of people would prefer to answer sensitive health questions posed by virtual assistants rather than people. Another study — this one by Juniper Research — found that chatbots could lead to cost savings of over $8 billion by 2020.One of the juggernauts looking to get in on the ground floor is none other than Microsoft. The Seattle company’s been experimenting with health care chatbots bots since 2017 — that’s the year Health Bot, a software-as-a-service offering within Microsoft’s Azure, made its debut. The idea was to provide a platform that’d enable care providers to deploy text, voice, and touch-response experiences capable of replying to questions with natural language, and that would integrate relatively seamlessly with Skype, Cortana, Facebook, Slack, SMS, and other mediums....