Microsoft Research

Posner Lecture: The Online Revolution: Learning without Limits - We are at the cusp of a major transformation in higher education.

A Kernel Test for Three-Variable Interactions - We introduce kernel nonparametric tests for Lancaster three-variable interaction and for total independence, using embeddings of signed measures into

NIPS Award Session Understanding Dropout - Dropout is a relatively new algorithm for training neural networks which relies on stochastically "dropping out” neurons during training in order to avoid

New Methods for the Analysis of Genome Variation Data - Genetic variation in genome sequences within a species such as humans underpins our biological diversity, is the basis for the genetic

IMS-Microsoft Research Workshop: Foundations of Data Science - Graph Cluster Randomization: Design and Analysis for Experiments in Networks A/B testing is a standard approach for evaluating the

Northwest Probability Seminar 2014 - Sharp higher order corrections for the critical value of a bootstrap percolation model Bootstrap Percolation (BP) models are simple cellular automata with a

From Bandits to Experts: A Tale of Domination and Independence - We consider the partial observability model for multi-armed bandits, introduced by Mannor and Shamir (2011).

MSR NYC Data Science Seminar Series: From "In" to "Over": Behavioral Experiments on Whole-Network Computation We report on a series of behavioral experiments in human computation on three different

Neural Reinforcement Learning - Reinforcement learning has become a wide and deep conduit that links ideas and results in computer science, statistics, control theory and economics to psychological

IMS-Microsoft Research Workshop: Foundations of Data Science - The small clustering problem: When the cluster sizes don't grow with the data Information about social entities is often spread across

Advances in Quantum Algorithms and Devices: A quantum algorithm for computing the unit group of an arbitrary degree number field

The traditional aim of machine learning methods is to infer meaningful features of an underlying probability distribution from samples drawn of that distribution.

The Origins of Common Sense: Modeling human intelligence with Probabilistic Programs and Program Induction

In this talk I will present new inference tools for adaptive statistical procedures.

This tutorial will look at how deep learning methods can be applied to problems in computer vision, most notably object recognition.

Machine Learning Algorithms Workshop: Logarithmic Time Online Multiclass Prediction & Log-Concave Sampling with SGD Logarithmic Time Online Multiclass prediction: We study the problem of multiclass

Joint Talk: Automatic Statistician - the big picture & Automatic construction and natural language description of nonparametric regression models

MSR NYC Data Science Seminar Series #4 - What Makes us Human?

Quantifiers meet their match(ing loop): new techniques and tools for dealing with unpredictable performance in Dafny Large developments using the Dafny program verifier often suffer from the

Microsoft Research Gender Diversity Lecture Series 4: Diversity Driving Innovation Moving from Research to Action

Microsoft Research – 7 июля 2016, 22:57

Spatially Defined Measures of Mobility from Call Data Records: Developing Big Data Measurements and Applications for Social Science Censuses and surveys have been the primary source of information

Algorithms increasingly run our lives. They find books, movies, jobs, and dates for us, manage our investments, and discover new drugs.

The authors of the bestselling Competing on Internet Time (a Business Week top 10 book) analyze the strategies, principles, and skills of three of the most successful and influential figures in

Researchers have found that the harried pace of modern office life is taking its toll on productivity, employee engagement, creativity and well-being.

Asymptotic Behavior of the Eden Model with Positively Homogeneous Edge Weights (a.k.a.

Completeness and incompleteness in Abstract Interpretation: A model for information concealment in code protection We will consider completeness and incompleteness of abstract interpretation as

Research community has done a significant amount of work on authenticating users based upon their typing behavior.

Computational Microscopy: Interactive Machine-Learning for Brain Activity Mapping and Understanding High-Dimensional Systems High-dimensional or complex systems, such as the brain, fundamentally

Scientists from Microsoft and the University of Washington have reached an early but important milestone in synthetic DNA storage by storing 200 megabytes of data on the molecular strands.