Microsoft Research

M. Lopes A Residual Bootstrap for High-Dimensional Regression with Near Low-Rank Designs B. McWilliams, G. Krummenacher, M. Lucic, J.

This tutorial offers a unified introduction to the modern theory of causality based on counterfactuals (aka potential outcomes), directed acyclic graphs (DAGs) and non-parametric structural equation

Combinatorial Pure Exploration of Multi-Armed Bandits We study the {\em combinatorial pure exploration (CPE)} problem in the stochastic multi-armed bandit setting, where a learner explores a set of

Role of Coupled Networks in 21st Century Energy Infrastructure ΓÇïOur modern economy overwhelmingly depends on the electricity infrastructure or the grid.

How transferable are features in deep neural networks?

Sparse Polynomial Learning and Graph Sketching Let f:{−1,1}n→R be a polynomial with at most s non-zero real coefficients.

J. Mairal, P. Koniusz, Z. Harchaoui, C. Schmid Convolutional Kernel Networks B. Zhou, A. Lapedriza, J. Xiao, A. Torralba, A. Oliva Learning Deep Features for Scene Recognition using Places Database W.

Sequence to Sequence Learning with Neural Networks Deep Neural Networks (DNNs) are powerful models that have achieved excellent performance on difficult learning tasks.

Learning Generative Models with Visual Attention Attention has long been proposed by psychologists to be important for efficiently dealing with the massive amounts of sensory stimulus in the

Feedforward Learning of Mixture Models We develop a biologically-plausible learning rule that provably converges to the class means of general mixture models.

Provable Submodular Minimization using Wolfe's Algorithm Owing to several applications in large scale learning and vision problems, fast submodular function minimization (SFM) has become a critical

Conditional Random Field Autoencoders for Unsupervised Structured Prediction We introduce a framework for unsupervised learning of structured predictors with overlapping, global features.

Asymmetric LSH (ALSH) for Sublinear Time Maximum Inner Product Search (MIPS) ΓÇïWe present the first provably sublinear time hashing algorithm for approximate \emph{Maximum Inner Product Search}

Median Selection Subset Aggregation for Parallel Inference ΓÇïFor massive data sets, efficient computation commonly relies on distributed algorithms that store and process subsets of the data on

Despite the scientific consensus on climate change, drastic uncertainties remain.

Analog Memories in a Balanced Rate-Based Network of E-I Neurons The persistent and graded activity often observed in cortical circuits is sometimes seen as a signature of autoassociative retrieval

P. Putzky, F. Franzen, G. Bassetto, J. Macke A Bayesian model for identifying hierarchically organised states in neural population activity L. Buesing, T. Machado, J. Cunningham, L.

G. Bresler, G. Chen, D. Shah A Latent Source Model for Online Collaborative Filtering S. Lim, Y. Chen, H. Xu Clustering from Labels and Time-Varying Graphs W. Liu, C. Mu, S. Kumar, S.

W. Su, S. Boyd, E. Candes A Differential Equation for Modeling NesterovΓÇÖs Accelerated Gradient Method: Theory and Insights V. Srikumar, C.

People increasingly are consuming live sports with a second screen in their hands -- a phone, a tablet, or a computer -- because they need more information to stay focused.

We have built a system that would enable the tracking of physical browsing by users in indoor spaces such as retail stores.

Larry Larsen speaks with Doug Burger about the findings of a Microsoft Research and Bing research project that equipped servers with reconfigurable hardware, in the form of field-programmable gate

Haptic Feedback

Image Composite Editor 2.0 introduces new features to the popular advanced panoramic image stitcher, including a new automatic image completion feature, which fills in the missing pixels when

Having made its debut at UIST 2014, RoomAlive uses multiple projectors to adapt interactive content that makes use of all of a roomΓÇÖs walls.

Senior Researcher Krysta Svore comments on the potential effect quantum computing can have on machine learning.

Limpopo TV White Spaces, a Microsoft 4Afrika initiative in partnership with the Department of Science and Technology, CSIR, the University of Limpopo and Multisource, in addition to education

Trishul Chilimbi, Partner Research Manager for Microsoft Research, discusses Project Adam, and how deep neural networks have enabled large-scale computer image recognition with astounding accuracy.

Demonstration of Microsoft Research Cambridge project, Digital Possessions.

Xim 1.3 now allows you to share your photos on the big screen using Xbox One. Available December 17 in the Windows Phone Store. getxim.com