Daniela Hernandez of Wired reports, “Drawing on the work of a clever cadre of academic researchers, the biggest names in tech—including Google, Facebook, Microsoft, and Apple—are embracing a more powerful form of AI known as ‘deep learning,’ using it to improve everything from speech recognition and language translation to computer vision, the ability to identify images without human help. In this new AI order, the general assumption is that Google is out in front… But now, Microsoft’s research arm says it has achieved new records with a deep learning system it calls Adam, which will be publicly discussed for the first time during an academic summit this morning at the company’s Redmond, Washington headquarters.” Read more
Posts Tagged ‘Microsoft’
Pedro Hernandez of eWeek reports, “‘Mobile-first, cloud-first’ may be Microsoft’s new mantra, but another term has been has been increasingly creeping into the company’s lexicon of late. As one of the components of Microsoft’s growing slate of smart services, machine learning is also guiding part of the company’s product strategy, according to Microsoft Research Distinguished Scientist John Platt. First, it helps to know how his company classifies machine learning (ML). ‘In general, ML converts data sets into pieces of software, known as ‘models,’ that can represent the data set and generalize to make predictions on new data,’ explained Platt in Microsoft’s new Machine Learning Blog.” Read more
Phil Goldstein of FierceWireless reported that, “Microsoft (NASDAQ: MSFT) joined the AllSeen Alliance, an open-source project founded on Qualcomm technology and aimed at coming up with a standard to connect devices and have them interact as part of the Internet of Things. The software giant’s participation in the group adds heft to its membership, which has been largely dominated by consumer electronics and home appliance makers.The AllSeen Alliance’s leading members include Haier, LG Electronics, Panasonic and Sharp, and in total the group now has 51 members. Adding Microsoft could ensure that future Windows devices interact with other connected gadgets via the AllSeen Alliance’s specifications.”
Microsoft is looking for a Senior Natural Language Processing Scientist in Redmond, WA. The post states, “Statistical Machine Translation (SMT) is one of the most interesting and best-known applications of Natural Language Processing and Machine Learning. The Microsoft Translator team is a very strong technical team that builds and runs the one of the largest automatic translation services on the Web, and is now expanding into speech translation with the Skype Translator app. We are looking for an experienced Research Scientist to join our team in working towards human quality text and speech translations. We are a small and dedicated group, focused on delivering the best quality translations for our users and internal and external partners. Entrepreneurship, passion, and self-drive are important qualities for the candidate to succeed in the team.” Read more
As The Semantic Web Blog discussed yesterday here, the Virtual Personal Assistant is getting more personal. Microsoft officially unveiled Cortana as part of the Windows Phone 8.1 smartphone software at its Build event yesterday, and the service effectively replaces the search function on Windows smartphones, both for the Internet and locally.
This statement served as the theme from corporate vice president and manager Joe Belfiore: “Cortana is the first truly personal digital assistant who learns about me and about the things that matter to me most and the people that matter to me most, that understands the Internet and is great at helping me get things done.”
The Bing-powered Cortana is launching in beta mode, and was still subject to a few hiccups during the presentation. For example, when Belfiore asked Cortana to give him the weather in Las Vegas, it reported the information in degrees, and was able to respond to his request to provide the same information in Celsius. But he couldn’t get her to make the calculations to Kelvin. But, he promised attendees, “Try it yourself because she is smart enough to tell you the answer in Kelvin.”
Juan Carlos Perez of InfoWorld reports, “Microsoft will add new software, developer tools and capabilities to Office 365 in an attempt to make the cloud applications suite a ‘smarter’ product that is better at helping people interact at work. At its SharePoint Conference, which kicks off in Las Vegas on Monday, Microsoft will demonstrate a new machine learning application code-named Oslo designed to understand how employees work in Office 365 and with whom. Oslo will base its insights on a variety of signals gleaned from how people use Office 365′s components, like Exchange Online for email, OneDrive for Business for storage, Lync Online for IM and video conferencing, SharePoint Online for team collaboration and Yammer for enterprise social networking. Microsoft calls this information the Office Graph.” Read more
Microsoft – as you’ve no doubt heard by now – has a new CEO. Satya Nadella most recently was Microsoft’s executive VP, cloud and enterprise group. But before that, the man who succeeds Steve Ballmer, he was senior vp, R&D, of online services and before that, the senior vp of search, portal and advertising group. Nadella has been at the company since 1992.
The man who succeeds Steve Ballmer has been referred to as the King of Bing, rebranding the search service from Live Search to Bing and getting kudos for making technical fixes. Announcing his promotion to president of Microsoft’s Server and Tools Business in 2011, Ballmer wrote in a memo that Nadella “led the overall R&D efforts for some of the largest online services and drove the technical vision and strategy for several important milestones, including the critical launch of Bing, new releases of MSN, Yahoo! integration across Bing and adCenter, and much more.”
Google’s letting the cash flow. Fresh off its $3.2 billion acquisition of “conscious home” company Nest, which makes the Nest Learning Thermostat and Protect smoke and carbon monoxide detector, it’s spending some comparative pocket change — $400 million – on artificial intelligence startup DeepMind Technologies.
The news was first reported at re/code here, where one source describes DeepMind as “the last large independent company with a strong focus on artificial intelligence.” The London startup, funded by Founders Fund, was founded by Demis Hassabis, Shane Legg and Mustafa Suleyman, with the stated goal of combining machine learning techniques and neuroscience to build powerful general purpose learning algorithms.
Its web page notes that its first commercial applications are in simulations, e-commerce and games, and this posting for a part-time paid computer science internship from this past summer casts it as “a world-class machine learning research company that specializes in developing cutting edge algorithms to power massively disruptive new consumer products.”
Bing is looking for a Senior Software Development Engineer in Bellevue, WA. The post states, “The Bing Local Search Relevance Platform team is responsible for building the world best relevance platform to serve the user’s search intent regarding location, business, address on web, mobile and map entry points and ensure the market expansion for Bing local search faster and cheaper crossing the world. Location and local query understanding is an important part of local search platform and we are looking for the talent who is interested in solving hard relevance problem in the scalable way: mining against large volume of data from web and internal logs; building scalable solution to construct the machine learned query intent classifiers and query parsers and other ranking features based on query and query context; handling the requests from markets about localization experience and design the platform feature to support the search relevance improvements for different cultures.” Read more
The media has been reporting the last few hours on the Obama administration’s self-imposed deadline for fixing HealthCare.gov. According to these reports, the site is now working more than 90 percent of the time, up from 40 percent in October; that pages on the website are loading in less than a second, down from about eight; that 50,000 people can simultaneously use the site and that it supports 800,000 visitors a day; and page-load failures are down to under 1 percent.
There’s also word, however, that while the front-end may be improved, there are still problems on the back-end. Insurance companies continue to complain they aren’t getting information correctly to support signups. “The key question,” according to CBS News reporter John Dickerson this morning, “is whether that link between the information coming from the website getting to the insurance company – if that link is not strong, people are not getting what was originally promised in the entire process.” If insurance companies aren’t getting the right information for processing plan enrollments, individuals going to the doctor’s after January 1 may find that they aren’t, in fact, covered.
Jeffrey Zients, the man spearheading the website fix, at the end of November did point out that work remains to be done on the backend for tasks such as coordinating payments and application information with insurance companies. Plans are for that to be in effect by mid-January.
As it turns out, among components of its backend technology, according to this report in the NY Times, is the MarkLogic Enterprise NoSQL database, which in its recent Version 7 release also added the ability to store and query data in RDF format using SPARQL syntax.
NEXT PAGE >>