Inside Microsoft Research

Sharing a Glimpse of the Future of Technology

Inside Microsoft Research — Tue, 21 May 2013 17:00:00 GMT

Posted by Rob Knies

It’s not often that people get a chance to peek into the future, but that will be the case May 21 in Washington, D.C., when Microsoft Research hosts its biennial D.C. TechFair.

During an afternoon open house held at the Microsoft Innovation & Policy Center in downtown Washington, D.C., customers, academia, and governmental officials will get an opportunity to explore the trends and technologies Microsoft Research expects to change the face of computing.

World-class scientists from Microsoft Research will demonstrate how new discoveries in computer science and information technology are not only enhancing Microsoft products but also helping to overcome some of society’s biggest challenges.

...(read more)

Watts Named A.D. White Professor-at-Large at Cornell

Inside Microsoft Research — Wed, 15 May 2013 16:00:00 GMT

Posted by Rob Knies

John Cleese, the acclaimed Monty Python actor, spent time as an A.D. White Professor-at-Large. So did renowned primatologist Jane Goodall. And Oliver Sacks, noted author and neurologist. And epic novelist Toni Morrison. And short-story writer Eudora Welty.

Add to that esteemed list the name of Duncan Watts, principal researcher at Microsoft Research New York City. On May 6, Watts’ name joined the roster of honorees for that Cornell University program, established in 1965 in conjunction with the university’s centenary and named for the school’s co-founder and first president, Andrew Dickson White.

“The A.D. White program brings some of the world’s most distinguished scholars, thinkers, and artists to Cornell as ‘professors-at-large,’” said Steven H Strogatz, Jac

...(read more)

A Trekkie’s Dream Come True

Inside Microsoft Research — Wed, 15 May 2013 01:43:00 GMT

Posted by Rob Knies

Do you speak Klingon? If not, that could all be about to change—thanks to Bing Translator’s just-released Klingon machine-translation system, developed in part by Microsoft Research.

For more details, see the post over at the Bing Translator Team Blog.

The language, familiar worldwide to numerous fans of the iconic science-fiction TV series Star Trek, has been added to the robust offerings already available on Bing Translator, just in time for the May 16 premiere of the new film Star Trek into Darkness.

...(read more)

Kati London of FUSE: Playful and Surprising

Inside Microsoft Research — Tue, 07 May 2013 16:24:00 GMT

Posted by Rob Knies

Many people talk to their plants. But what if those plants were able to talk back?

That’s the premise behind Botanicalls, a project to enable communications between plants and people. A sensor network provides the flora the ability to call and text people to request assistance, such as “I need water,” or “Not sure if it was you, but someone gave me a drink—I feel great!”

It’s a fascinating, precocious venture, one featured on the TODAY show on May 7 as part of a discussion about home technologies. It’s also a window into the work of Kati London, one of the driving forces behind Botanicalls and the newest member of FUSE Labs at Microsoft Research.

...(read more)

Putting Your Work ID Badge to Work

Inside Microsoft Research — Wed, 01 May 2013 17:19:00 GMT

Posted by Rob Knies

Steve Hodges and his colleagues in the Sensors and Devices group at Microsoft Research Cambridge spend their time pursuing novel sensing technologies and new devices that make it easier for people to interact with computer systems and digital content.

The team’s successes have been many, and among the most notable have been SenseCam—a wearable camera that takes photos automatically, thereby enabling users to review a series of snapshots and recall events as they transpired—and .NET Gadgeteer, a rapid prototyping platform for small electronic gadgets and embedded hardware devices.

Now, these creative researchers have unveiled their latest concept via a note titled An Interactive Belt-worn Badge with a Retractable String-based Input Mechanism during the Association for Computing Machinery’s 2013 SIGCHI Conference on Human Factors in Computing Systems, being held in Paris through May 2.

The note was co-written by Norman Pohl of the University of Stuttgart, Hodges and his Microsoft Research Cambridge colleagues Nicolas Villar and John Helmes, along with Tim Paek of Microsoft Research Redmond.

“We interact with digital content more and more—such as electronic diaries, emails, traffic status, and weather info,” says Hodges, principal hardware engineer at the Cambridge lab. “But even if you have your mobile phone in your pocket, it can be a pain to interact with this content in many cases.

“The badge is always on hand and lets you navigate to the content you want simply by moving it to the right place relative to your body, using your spatial muscle memory. It’s much easier for quick ‘snacking’ on small amounts of digital content.”

The lightweight, interactive badge prototype includes an embedded LCD that presents dynamic information to the wearer. Sensing-based input capabilities are built into the badge’s retractable string, enabling single-hand interaction.

“When you pull the badge away from your body,” Hodges explains, “the sensor detects how far the badge is pulled out and at what angle, enabling the system to know where the badge is in relation to your body. Depending on this location, it displays different content. If the content to be displayed is too big to fit on the badge display, it’s possible to pan around it.”

Paek, in particular, served as a catalyst for moving the research project to where it is today.

“Our work on augmented-reality systems—like HoloDesk and the mobile projector,” Hodges says, “got us thinking about a lightweight display which could sense its location relative to the body and which could act like a ‘lens’ onto virtual digital content.

“But it was a conversation with Tim that actually spurred us to turn all these thoughts into a full-fledged research project when we realized that his vision about displaying automatically mined context information on a low-power, wearable device like a badge were similar to ours.”

Of course, the exploration of a new, intriguing research project has its own attractions.

“This research still has many open questions on design, form factor, and content to be explored, but the challenge of working on technology that has promising, unexplored potential is what makes this exciting,” he enthused. “So many people already wear a badge on a regular basis—in offices, hospitals, or schools and universities—yet they are currently just pieces of plastic with static images on them.

“Let’s turn them into interactive devices!”

Helping the Low-Literate Learn to Navigate Through User Interfaces

Inside Microsoft Research — Tue, 30 Apr 2013 07:01:00 GMT

Posted by Rob Knies

For several years, researchers from Microsoft Research India’s Technology for Emerging Markets (TEM) group have been studying how to design applications for economically poor communities such as those found in India.

In particular, Indrani Medhi, a researcher at the India lab, has been focusing on user interfaces for low-literate and novice technology users. Medhi, who is completing her Ph.D. at the Industrial Design Centre at the Indian Institute of Technology Bombay, has co-written a paper accepted for the Association for Computing Machinery’s 2013 SIGCHI Conference on Human Factors in Computing Systems (CHI). The paper is titled Some Evidence for the Impact of Limited Education on Hierarchical User Interface Navigation and was written with Meera Lakshmanan, a translator and research assistant; Kentaro Toyama, a former head of TEM and now a researcher at the University of California, Berkeley and a fellow of the Dalai Lama Center for Ethics and Transformative Values at the Massachusetts Institute of Technology; and Edward Cutrell, Toyama’s successor as senior research manager of TEM.

The paper examines one factor in application design for poor communities: the fact that users with little or no education have a diminished capacity to navigate a hierarchical user interface. Medhi’s work has explored ways that UIs can be designed for low-literate people by using text-free iconography that such users can recognize, but the challenge continues.

...(read more)

Pursuing Insights About Health and Well-Being from Social Media

Inside Microsoft Research — Thu, 25 Apr 2013 16:48:00 GMT

Posted by Eric Horvitz and Munmun De Choudhury

At Microsoft Research, we’ve been exploring the use of data analysis and machine learning to gain insights about health and well-being—and to enhance the quality of health care. Our efforts in this area include research on using data stored in electronic health records to construct predictive models that can provide physicians with advance warning about patient outcomes.

We’ve worked with colleagues to develop systems that can predict the likelihood that a patient will contract an infection while in the hospital or that a patient being discharged will be readmitted to the hospital within a short time. Some of these models have been deployed and are in use at hospitals throughout the world, providing demonstrated value to patients and physicians.

Beyond examining data from medical health records about hospitalized patients, we have been interested in the prospects of developing new methods that can transform anonymized data about the search and communications activities of people into a large-scale sensor network for public health. As an example of directions and opportunities in this realm, we recently showed how we can detect previously unknown drug interactions via analysis of anonymized web-search logs. We identified useful signals via analysis of tens of millions of queries sent to search engines by millions of users who had consented to share their search activities with Microsoft for research purposes.

...(read more)

Improved Healthcare via Machine Learning: a Way Forward

Inside Microsoft Research — Wed, 24 Apr 2013 21:10:00 GMT

Posted by Rob Knies

The Microsoft Research Machine Learning Summit 2013 concluded with a plenary panel discussion titled Data Challenges and Opportunities in the Next Decade. Chaired by Jeannette Wing, Microsoft vice president and head of Microsoft Research International, the discussion included Eric Horvitz, Microsoft distinguished scientist and managing co-director of Microsoft Research Redmond; Michel Cosnard, president of Inria; Iain Buchan of the University of Manchester; and Lionel Tarassenko of the University of Oxford.

My previous post ended with Hermann Hauser, co-founder of Amadeus Capital Partners, stating that machine learning would have a profound effect on the future of health care. That was interesting, because I had planned for the final post from the summit to focus on that very subject.

Buchan is quite aware of that potential. A clinical professor of Public Health Informatics at the University of Manchester and director of the MRC Health eResearch Centre, his research interests lie in building effective models of health and in connecting patients and health professionals with more potent health information.

...(read more)

The Sixth Wave Is upon Us

Inside Microsoft Research — Wed, 24 Apr 2013 14:48:00 GMT

Posted by Rob Knies

The second day of the Microsoft Research Machine Learning Summit 2013 got off to a rousing start with an hour-long plenary keynote by serial entrepreneur Hermann Hauser, co-founder of Amadeus Capital Partners.

Hauser, a physicist and a Fellow of the Royal Society, the Institute of Physics, and the Royal Academy of Engineering, has a long, successful history of in incubating IT companies, including U.K. computer maker Acorn Computers, a former subsidiary of which is now known as ARM Holdings, which dominates the market for chips used in mobile phones.

His talk was called Machine Learning, the 6th Wave of Computing, and he began by referring back to 1947 and the Electronic Delay Storage Automatic Calculator (EDSAC), a prototype computer constructed by British computing pioneer Maurice Wilkes at the University of Cambridge.

Hauser referred to the EDSAC as the “0th Wave” and used that as his springboard into a discussion of the evolution of the computer from early days up to the present. The six waves, Hauser said, consist of:

The mainframe.
The mini-computer.
The workstation.
The personal computer.
The smartphone combined with the cloud.
Ubiquitous computing and machine learning.

For each, he discussed associated developments, such as the hardware that enabled the next leap forward, the user interface, and how the computer. He also noted that each wave led to a mushrooming number of devices, while the costs of the computing devices were driven steadily down. Then he got to our current era.

“Interesting point about the price,” Hauser said. “I told you that phones are available at zero cost, really, so when you go from there, you can’t really go below zero—but in a way you can.

“The thing that has been happening with the price in the Internet of Things in the ubiquitous-computing wave is that the price actually is no longer associated with the computing element. The price is a thing that you’d buy anyway, like a car or glasses that you need or a thermostat. It’s just that now, it has these nice, new functionalities that make it really useful. It becomes part of the environment.”

And then, Hauser tied his historical recap back to the subject of the event.

“What’s the big deal about machine learning?” he asked. “It really makes the Internet, the everyday things you interact with, much more user-friendly, much more human.

“But the big thing is the effect this will have on health.”

That comment provided an ultra-smooth segue into my next post. Stay tuned.

Infer.NET: Machine Learning Tailor-Made

Inside Microsoft Research — Tue, 23 Apr 2013 18:06:00 GMT

Posted by Rob Knies

One of the featured technologies on display on April 23, the first day of Microsoft Research Machine Learning Summit 2013, was Infer.NET, a powerful, compelling .NET library from Microsoft Research Cambridge.

Infer.NET is an example of model-based machine learning, as explained by Tom Minka from the Cambridge lab during a morning talk.

“It’s about trying to get more people to try machine learning,” said Minka, a senior researcher. “The traditional approach to this is that experts build prepackaged learners that are very generic and apply in a robust way to different data sets. But the problem with that approach is that it doesn’t account for domain knowledge. In lots of areas where we want to use machine learning, such as vision or speech or ecology, there is very strong domain knowledge.

...(read more)