In this paper, we propose Partition min-Hash (PmH), a novel hashing scheme for discovering partial duplicate images from a large database. Unlike the standard min-Hash algorithm that assumes a bag of words image representation, our approach utilizes the fact that duplicate regions among images are often localized. By theoretical analysis, .. more »

This paper reports our recent exploration of the layer-by-layer learning strategy for training a multi-layer generative model of patches of speech spectrograms. The top layer of the generative model learns binary codes that can be used for efficient compression of speech and could also be used for scalable speech recognition or rapid speech content .. more »

We extend our earlier work on deep-structured conditional random field (DCRF) and develop deep-structured hidden conditional random field (DHCRF). We investigate the use of this new sequential deep-learning model for phonetic recognition. DHCRF is a hierarchical model in which the final layer is a hidden conditional random field (HCRF) and the .. more »

The advance of GPS-enabled devices facilitates people to record their location histories with GPS traces, which imply human behaviors and preferences related to travel. In this paper, we perform two types of travel recommendations by mining multiple users’ GPS traces. The first is a generic one that recommends a user with top interesting lo.. more »

While WiFi-based indoor localization is attractive, the need for a significant degree of pre-deployment effort is a key challenge. In this paper, we ask the question: "Can we perform indoor localization with no pre-deployment effort?" Our setting is an indoor space, such as an office building or a mall, with WiFi coverage but where we do not assume .. more »

We introduce a new descriptor for images which allows the construction of efficient and compact classifiers with good accuracy on object category recognition. The descriptor is the output of a large number of weakly trained object category classifiers on the image. The trained categories are selected from an ontology of visual concepts, but the .. more »

Regression Forests for Efficient Anatomy Detection and Localization in CT Studies.. more »

Recently, Deep Belief Networks (DBNs) have been proposed for phone recognition and were found to achieve highly competitive performance. In the original DBNs, only frame-level information was used for training DBN weights while it has been known for long that sequential or full-sequence information can be helpful in improving speech recognition .. more »

RearType is a text input system for mobile devices such as Tablet PCs, using normal keyboard keys but on the reverse side of the device. The standard QWERTY layout is split and rotated so that hands gripping the device from either side have the usual keys under the fingers. This frees up the front of the device, maximizing the use of the display .. more »

We present a framework for vision-assisted tagging of personal photo collections using context. Whereas previous efforts mainly focus on tagging people, we develop a unified approach to jointly tag across multiple domains (specifically people, events, and locations). The heart of our approach is a generic probabilistic model of context that couples .. more »

Long-term personal GPS data is useful for many UbiComp services such as traffic monitoring and environmental impact assessment. However, inference attacks on such traces can reveal private information including home addresses and schedules. We asked 32 participants from 12 households to collect 2 months of GPS data, and showed it to them in .. more »

This is a .NET assembly with a PowerShell front end to enable interactive physical-design tuning sessions over SQL Server databases... more »

We consider the potential for network trace analysis while providing the guarantees of “differential privacy.” While differential privacy provably obscures the presence or absence of individual records in a dataset, it has two major limitations: analyses must (presently) be expressed in a higher level declarative language; and the analysis resu.. more »

Researchers from Microsoft Research are presenting seven papers during SIGCOMM 2010, being held in New Delhi, India, from Aug. 30 to Sept. 3... more »

In this paper, we present RePriv, a system for managing and controlling the release of private information from the browser. We demonstrate how always-on user interest mining can effectively infer user interests in a real browser. We go on to discuss an extension framework that allows third-party code to extract and disseminate more detailed .. more »

Five young people from Microsoft Research have been named 2010 recipients of the TR35 award, presented by Technology Review to the world’s most exciting young innovators... more »

A collection of tools and analyses designed for network-trace analysis using the Privacy Integrated Queries data-analysis framework... more »

This paper makes three significant extensions to a noisy channel speller designed for standard writ-ten text to target the challenging domain of search queries. First, the noisy channel model is sub-sumed by a more general ranker, which allows a variety of features to be easily incorporated. Se-cond, a distributed infrastructure is proposed for .. more »

The Microsoft Research Graduate Women’s Scholarship Program, a one-year scholarship program for outstanding women graduate students that supports women in the second year of their graduate studies, has begun accepting applications... more »

Internships with Microsoft Research Asia have fostered a passion for the value of collaborative relationships for recent University of Tokyo Ph.D. Yusuke Sugano... more »