Microsoft Research

This talk discards hand-wavy pop-science metaphors and answers a simple question: from a computer science perspective, how can a quantum computer outperform a classical computer?

A fundamental goal in data analysis is learning which actions (i.e. interventions) are optimal for producing a desired outcome within specific individuals or over an entire population.

As market design theory increasingly shapes the design and operations of real-life marketplaces, it is important for designers to provide simple policy levers that practitioners can use to optimize

This talk discusses the epistemic, societal, and ethical implications of making publicly available and reusing open data from biomedical research.

Episode 12 | February 15, 2018 If you’ve ever wondered if you could find the perfect combination of computer scientist… and MacGyver, look no further than Dr.

Azure Machine Learning Accelerated Models, powered by Project Brainwave, deliver real-time AI with phenomenal performance at affordable cost.

Microsoft is leading the industry’s datacenter transformation by using programmable hardware: First to prove the value of FPGAs for cloud computing, first to deploy them at cloud scale, and, with

The modern world is full of algorithms that use data to make decisions. The data often come from people; the predictions or decisions often affect people. Yet classically, the study of e.g.

This talk explores the concept of the connected self-owner, which takes account of the metaphysical significance of relations among persons for persons’ capacities to be owners.

By some accounts, JavaScript is the most popular programming language in the world. Yet, it’s also not a language that was designed with fast performance in mind.

Machine learning (ML) and automatic algorithmic decision making have started to play central and crucial roles in our daily lives.

Episode 21 | April 25, 2018 Dr.

Researchers at Microsoft strive to advance perhaps one of the most challenging areas of research and development in virtual reality.

Researcher Meredith Ringel Morris discusses three projects highlighted at CHI 2018 which focus on improving technology for people with low-vision or blindness.

Most deep reinforcement learning algorithms are data inefficient in complex and rich environments, limiting their applicability to many scenarios.

Many real-world tasks require understanding interactions between a set of entities.

The performance of stochastic gradient descent (SGD) can be improved by actively selecting mini-batches. In this work, we explore active mini-batch selection using repulsive point processes.

The importance-weighted autoencoder (IWAE) approach of Burda et al. (2015) defines a sequence of increasingly tighter bounds on the marginal likelihood of latent variable models.

Cancer is often viewed as a genetic disease because accumulation of mutations in the DNA are said to “drive” the progression of the disease.

This talk introduces variational continual learning, a simple but general framework for continual learning that fuses online variational inference (VI) and recent advances in Monte Carlo VI for

Much research in Bayesian Deep Learning is about approximating the posterior. With some notable exceptions, the choice of prior is less often considered.

Deep reinforcement learning (RL) has shown promising results for learning complex sequential decision-making behaviors in various environments.

Episode 20 | April 18, 2018 Emotions are fundamental to human interaction, but in a world where humans are increasingly interacting with AI systems, Dr.

The concept of an “uncanny valley”—whereby robots or avatars that exhibit more and more humanlike appearance and behavior can (perhaps counter-intuitively) lead to sharp reduction in the subjective

Episode 19 | April 11, 2018 From ancient hieroglyphics to secret decoder rings to World War II Enigma code-makers and code-breakers, cryptography has always held a particular fascination for us.

Fireside Chat between Eric Horvitz and Jon Kleinberg. See more at microsoft.com/en-us/research/videos/ai-distinguished-lecture-series

Fireside Chat between Eric Horvitz and Tom Dietterich. See more at microsoft.com/en-us/research/videos/ai-distinguished-lecture-series

Fireside Chat between Adith Swaminathan and Eva Tardos. See more at microsoft.com/en-us/research/videos/ai-distinguished-lecture-series

The distinguished Rochester Institute of Technology (RIT) in Rochester, New York, is renowned for graduating successful professionals who are deaf and hard of hearing.

We consider the problem of designing mechanisms for online marketplaces for long-lived agents with persistent preferences.