Microsoft Research

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.

The Neuroscience of Intelligence Yesterday’s scientific research, starting with Hubel and Wiesel’s Nobel-prize winning work on the circuitry underlying visual processing in cortex, gave rise to

1:00 pm - 2:30p - Tutorial session: n-gram and Neural Network Language Modeling, Ashish Vaswani

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.

The goal of this workshop is to inform people about open source machine learning systems being developed, aid the coordination of such projects, and discuss future plans.

Digits, a wrist-worn gloveless sensor developed by Microsoft Research in Cambridge, U.K., enables 3-D computer interaction in any environment and is practical beyond computer gaming.

A Join circuit implemented in DNA

Taking the real world example of ad placement on web search result pages, this talk (1) provides a real world example demonstrating the value of causal inference for large-scale interactive machine

Graph-structured data has become a universal phenomenon in the sciences, and novel mathematics is required to tackle the problems stemming from analyzing this data.

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.

Probabilistic modelling provides a mathematical framework for understanding what learning is, and has therefore emerged as one of the principal approaches for designing computer algorithms that

I envision a system that enables successful collaborations between humans and machine learning models by harnessing the relative strength to accomplish what neither can do alone.

Submodular and supermodular functions have found wide applicability in machine learning, capturing notions such as diversity and regularity, respectively.

In the talk, I will introduce a model of learning with Intelligent Teacher.

The goal of this workshop is to inform people about open source machine learning systems being developed, aid the coordination of such projects, and discuss future plans.

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 propose a technique for learning representations of parser states in transition-based dependency parsers.

Motivated by machine learning problems over large data sets and distributed optimization over networks, we consider the problem of minimizing the sum of a large number of convex component functions.

How Much Information Does a Human Translator Add to the Original and Multi-Source Neural Translation

Probabilistic programming has recently attracted much attention in Computer Science and Machine Learning communities.

Over the past few years, virtual reality has experienced a remarkable resurgence.

Machine Learning with text data can be very useful for social networks analytics for instance to perform sentiment analysis.

Many potential applications of graphical models (such as Conditional Random Fields) are complicated by the fact that exact inference is intractable.

Public interest in Machine Learning is mounting as the societal impacts of technologies derived from our community become evident.

The promise and scale of mobile money has ignited huge interest in digital finance products in emerging markets, and is genuinely transforming these countries by providing access to formal financial

Hallmarks of Deep Learning in the Brain Anatomically, the brain is deep. To understand the ramifications of depth on learning in the brain requires a clear theory of deep learning.

Analysis of medical images is essential in modern medicine.

Many people delay or completely avoid updating the software on their devices. This can cause serious issues for security and software management.

Locality-Sensitive Hashing (LSH) is a powerful technique for the approximate nearest neighbor search (ANN) in high dimensions. In this talk I will present two recent results.

k n (3+o(1))/n^{1/3} \tilde{\mathcal{O}}(\min(k/n,1/\sqrt n)) \ge\tilde\Omega(\min(k/n,1 n^{2/3})) \ge\tilde\Omega(\min(k/n,1/\sqrt{n})) Estimating distributions over large alphabets is a