Microsoft Research

The Local Search method was illustrated using three problems: Maximum cut (2-approximation), Maximum leaf spanning trees (10-approximation) and the metric k-median problem (3-approximation).

Microsoft Research – 12 августа 2016, 3:34

The lecture starts with an outline of the topics proposed to be covered, followed by an introduction to greedy algorithms illustrated using the set cover problem (logarithmic ratio approximation).

Microsoft Research – 12 августа 2016, 3:31

Two LP-based methods were covered. Filtering the optimal LP solution was illustrated to design a constant factor approximation for the metric uncapacitated facility location problem.

Microsoft Research – 12 августа 2016, 3:31

Getting started with big data? Generating more and more data without the hardware resources to process it?

Microsoft Research – 12 августа 2016, 3:29

Microsoft Research – 12 августа 2016, 3:26

Earth Observations and other environmental data collection methods help us accumulate terabytes to petabytes of datasets. This pose a grand challenge to the informatics for environmental studies.

S V N Vishwanathan (Vishy) and Prateek Jain will offer a 10 week Machine Learning course. It will be an exciting mix between theory and application talks by various guest lecturers.

An auction designer faces an optimization problem with a twist.

This presentation proposes a noise-robust speech recognition method composed of weak noise suppression (NS) and weak Vector Taylor Series Adaptation (VTSA).

S V N Vishwanathan (Vishy) and Prateek Jain will offer a 10 week Machine Learning course. It will be an exciting mix between theory and application talks by various guest lecturers.

Crowdsourcing sites like Amazon's Mechanical Turk are increasingly being used by researchers to construct 'virtual labs' in which they can conduct behavioral experiments.

The lecture starts with an outline of the topics proposed to be covered, followed by an introduction to greedy algorithms illustrated using the set cover problem (logarithmic ratio approximation).

Although mediated humor is pervasive in our media cultures, media studies have largely glossed over the role of technology in the process of making humor.

The prospect of using auctions to resolve conflicts among parties competing for the same top-level internet domains is described.

A basic fact in algebraic graph theory is that the number of connected components in an undirected graph is equal to the multiplicity of the eigenvalue zero in the Laplacian matrix of the graph.

Miku Hatsune is Japan's number one virtual idol.

The talk will be structured in two parts. I will first talk about my research on marker-less motion capture, in particular presenting my recent ECCV work on hands tracking.

Data Visualization in Virtual Spaces and High Dimensions Visualization is a bridge between the quantitative content of data and the human intuition and understanding.

S V N Vishwanathan (Vishy) and Prateek Jain will offer a 10 week Machine Learning course. It will be an exciting mix between theory and application talks by various guest lecturers.

We develop and analyze stochastic optimization algorithms for problems in which the expected loss is strongly convex, and the optimum is (approximately) sparse.

In less than a decade the internet has provided us access to enormous quantities of chemistry data. Chemists have embraced the web as a rich source of data and knowledge.

This session will explore the newly added C++ support in WP8 and the new application models where C++ is available.

Over 200 million people now trust an algorithm theyΓÇÖve never met to listen to and discover music.

S V N Vishwanathan (Vishy) and Prateek Jain will offer a 10 week Machine Learning course. It will be an exciting mix between theory and application talks by various guest lecturers.

Funders, researchers, and public stakeholders increasingly see the need to better communicate and curate ever expanding bodies of research data.

Machine Learning Work Shop-Session 4 - Dengyong Zhou - 'A Minimax Entropy Method for Learning the Wisdom of Crowds' In many real-world machine learning applications, we have training examples