Microsoft Research

Selecting an appropriate set of indexes to build in a database for a given workload can result in significant reductions in query execution cost, e.g., CPU time.

Nathalie Riche's short talk on Digital ink + Data at Microsoft PhD Summit 2019.

Ken Hinckley talks about practical approaches to sensing the “seen but unnoticed” vocabulary of natural touch and device interaction.

Professor Donald Kossmann, Distinguished Scientist and Director or the Redmond Lab gave the Microsoft PhD Summit opening keynote on the topic of The Global AI Supercomputer and the next era of

Models that map natural language (NL) to source code in general purpose languages such as Java, Python, and SQL find utility amongst two main audiences viz.

This video shows a demonstration of a research prototype that distinguishes hand and user on a large display with two users working side by side.

Episode 95 | October 16, 2019 Machine learning is a powerful tool that enables conversational agents to provide general question-answer services.

Reinforcement Learning Day 2019: Generalization in Reinforcement Learning with Selective Noise Injection See more at

We explore a future in which people spend considerably more time in virtual reality, even during moments when they walk between locations in the real world.

We are motivated by the problem of building autonomous robots that are able to perform complex tasks ranging from inspection and diagnosis to manipulating soft tissue.

In recent years, a technology known as Remote Direct Memory Access (RDMA) has made its way into data centers, earning a spotlight in distributed systems research.

Creating or arranging objects at runtime is needed in many virtual reality applications, but such changes are noticed when they occur inside the user’s field of view.

We introduce CapstanCrunch, a force resisting, palmgrounded haptic controller that renders haptic feedback for touching and grasping both rigid and compliant objects in a VR environment.

We explore a future in which people spend considerably more time in virtual reality, even during moments when they walk between locations in the real world.

Mainstream social computing technologies—like social media platforms, online discussion forums, or crowdsourcing marketplaces—have revolutionized how literate people with access to smartphones and

Episode 94 | October 16, 2019 Dr. Craig Costello is in the business of safeguarding your secrets. And he uses math to do it.

Reinforcement Learning Day 2019: Grounding Natural Language for Embodied Agents See more at microsoft.com/en-us/research/video/grounding-natural-language-for-embodied-agents

Reinforcement Learning Day 2019: Towards Using Batch Reinforcement Learning to Identify Treatment Options in Healthcare See more at

Reinforcement Learning Day 2019: Can quantum mechanics help us learn models of classical systems?

Reinforcement Learning Day 2019: Reinforcement Learning From Small Data in Feature Space See more at

Reinforcement Learning Day 2019: Scalable and Robust Multi-Agent Reinforcement Learning See more at

Reinforcement Learning Day 2019: Reward Machines: Structuring Reward Function Specifications and Reducing Sample Complexity in Reinforcement Learning See more at

Reinforcement Learning Day 2019: Safe and Fair Reinforcement Learning See more at microsoft.com/en-us/research/video/safe-and-fair-reinforcement-learning

The visual world around us is highly structured. As 2D projection of our world, images are also structured.

While the majority of hospital appointments in the ambulatory setting as well as General Practitioners use Electronic Health Record (EHR) systems, these are poorly designed for use during patient

HapSense: A Soft Haptic I/O Device with Uninterrupted Dual Functionalities of Force Sensing and Vibrotactile Actuation We present HapSense, a single-volume soft haptic I/O device with uninterrupted

Optics for the Cloud Talks: Advanced polarized light microscopy for mapping molecular orientation Polarization is a basic property of light, but the human eye is not sensitive to it.

Episode 93 |

Current statistical tools place the burden of valid, reproducible statistical analyses on the user.

Distributed systems are hard to get right.