Microsoft Research

The era of increasing single-threaded application performance is long over. Application code bases must adapt to utilize multicore chips, or be forever stuck with circa 2004 performance levels.

As people are increasingly connected to other people and computer agents, forming mixed networks, collaborative teamwork offers great promise for transforming the way people perform their everyday

MAP inference in MRFs (or energy minimization) is known to be NP-hard in general, and thus research has focussed on either finding efficiently solvable subclasses (eg trees via BP), or approximate

The lines between mobile devices - such as PDAs, mobile phones and cameras - have blurred, and are now almost gone.

The problem of dimension reduction has inspired many different methods over the years in the statistics and machine learning communities.

The aim of this interactive session is to deconstruct operations in a high-volume, urban Emergency Department, examining the concepts of patient and information flow and how they relate to medical

We will describe recent advances in deep learning techniques for Natural Language Processing (NLP).

For centuries, the general view had been that science is neither good nor bad, that it merely supplies information ad that new information is always beneficial.

The Advanced Encryption Standard (AES) is one of the most popular ciphers in the world and is widely used for both commercial and government purposes.

With the advent of the Internet it is now possible to collect hundreds of millions of images. These images come with varying degrees of label information.

Testing to see if a Boolean circuit computes the identically zero function is a fundamental problem in computational complexity.

There are companies that create waves and those that ride or are drowned by them. LetΓÇÖs take a ride on the Google wave, from how it formed to how it crashed into traditional media businesses.

It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard for their own interest. Each participant in a competitive economy is led

Automated decision procedures nowadays play a crucial role in the verification process of real life code.

AES is the best known and most widely used block cipher.

The human object recognition system has about a trillion synapses and computer vision systems will probably need to learn a similar number of parameters in order to be competitive.

The shift from single to multiple core architectures means that programmers must write concurrent, multithreaded programs in order to increase application performance.

The obvious way to find images that are semantically similar to a query image is to solve the object recognition problem.

An overview on recent work on map labeling will be presented with emphasis on labeling with leaders.

Accessible computing refers to software and hardware solutions that make computers and other aspects of life more accessible to persons with disabilities.

Constraint programming (CP) is a general and powerful method to solve some combinatorial problems.

We present a technique for designing memory-bound algorithms with high data reuse on Graphics Processing Units (GPUs) equipped with close-to-ALU software-managed memory.

Many large-scale distributed environments, aka ``clouds'', ``grids'', or ``batch systems'', execute Bags-of-(millions of) Tasks (BOTs).

Climate scientists build large, complex simulations with little or no software engineering training, and do not readily adopt the latest software engineering tools and techniques.

The World Wide Web has changed almost every aspect of modern life and touches us all. We use it to shop, date, entertain, communicate and research.

The trend towards multi-core computing has made concurrent software an important target of computer-aided verification.

COMPANIONS is an EU project that aims to change the way we think about the relationships of people to computers and the Internet by developing a virtual conversational 'Companion'.

This talk addresses the question of whether message passing provides an adequate programming model for multicore processors.

In this talk I will discuss research we have been conducting on the recognition of engagement and a re-usable module for engagement recognition in human-robot dialogs.

Scripting languages are playing an increasing role in todayΓÇÖs software landscape due to their support for rapid and exploratory development.