Microsoft Technology Licensing, LLC

Elwha LLC

Other

Elwah LLC

Mircrosoft Corporation

Microsoft Corproration

Eric J Horvitz

Carl Myers Kadie

Gary William Flake

John Charles Krumm

Susan Theresa Dumais

Joshua T Goodman

Johnson T Apacible

Lili Cheng

Raman Kumar Sarin

Oliver Hurst-Hiller

Alexander G Gounares

Ryen William White

Chris Demetrios Karkanias

William Henry Gates, Iii

Christopher A Meek

Lowell L Wood, Jr

Roderick A Hyde

Clarence T Tegreene

Edward K Y Jung

Nathan P Myhrvold

Charles Whitmer

Douglas Christopher Burger

Craig J Mundie

Eric David Brill

Jordin T Kare

Eric Leuthardt

Kenneth Paul Hinckley

Xuedong D Huang

Desney S Tan

Richard Neil Zare

Feng Zhao

Daniel Chaim Robbins

Kamal Jain

Ashish Kapoor

Anoop Gupta

Raymond Edward Ozzie

David Maxwell Chickering

Timothy S Paek

Curtis Glenn Wong

Mark H Lucovsky

Shaun Douglas Pierce

Michael Jack Sinclair

Gur Kimchi

Surajit J Chaudhuri

Ranveer Chandra

Hrvoje Benko

John Carlton Platt

George G Robertson

Mary P Czerwinski

Milind V Mahajan

Michael J Connolly

Paul Nathan Bennett

Todd E Holmdahl

Jie Jerry Liu

Andrew David Wilson

William Jefferson Westerinen

Aman Kansal

Gregory Ray Smith

Kenneth D Ray

Sridhar Srinivasan

Marc Eliot Davis

Stephen Edward Hodges

Shahram Izadi

Mehmet Iyigun

Matthew Graham Dyor

James William Scott

Kentaro Toyama

Patrick Markus Baudisch

Brian R Meyers

Krysta Marie Svore

Alexander Frank

David Alexander Butler

Roger Shane Barga

Stephen L Lawler

Dale Alan Sather

Kenneth S Reneris

Hubert Van Hoof

Alistair K Chan

Alejandro Acero

Heung-Yeung Shum

Eyal Ofek

Paul England

Jin Li

Henrique Sarmento Malvar

Shai Guday

Arun Upadhyaya Kishan

Chaitanya Dev Sareen

Christian L Belady

Dawson Yee

Hari R Pulapaka

Michael Richard Fortin

Paul A Viola

Bongshin Lee

Rene Antonio Vega

Todd D Newman

Brett David Brewer

Zhaowei Charlie Jiang

Michael Gamon

Melissa Wood Dunn

George P Moromisato

The technology described herein builds an optimal refresh schedule by minimizing a cost function constrained by an available refresh bandwidth. The cost function receives an importance score for a content item and a change rate for the content item as input in order to optimize the schedule. The cost function is considered optimized when a refresh schedule is found that minimizes the cost while using the available bandwidth and no more. The technology can build an optimized schedule to refresh content with incomplete change data, content with complete change data, or a mixture of content with and without complete change data. It can also re-learn content item change rates from its own schedule execution history and re-compute the refresh schedule, ensuring that this schedule takes into account the latest trends in content item updates.

Efficient freshness crawl scheduling

Systems and methods to determine when a media is a high-fidelity reproduction of an original media from a trusted entity are disclosed. In certain aspects, systems and method for generating a fragile watermark are disclosed. The fragile watermark may be inserted into digital media in a manner such that the watermark cannot be identified if the media content is significantly altered. Media content may be subsequently analyzed to determine the presence of a fragile watermark. When the fragile watermark is present, provenance of the media content can be verified and an indication of provenance is provided to the user.

Media provenance certification via fragile watermarking

Biases in search and retrieval (i.e., situations where searchers seek or are presented with information that significantly deviates from the truth) may be detected by comparison to one or more authoritative sources. Once bias or potential bias is detected, techniques may be applied to indicate and/or compensate for the bias. Such techniques may allow users to more easily assess the veracity of search results, and increase the chances that users will locate accurate answers to their questions.

Compensating for bias in search results

The present disclosure relates to systems, methods, and computer readable media that evaluate performance of a machine learning system in connection with a test dataset. For example, systems disclosed herein may receive a test dataset and identify label information for the test dataset including feature information and ground truth data. The systems disclosed herein can compare the ground truth data and outputs generated by a machine learning system to evaluate performance of the machine learning system with respect to the test dataset. The systems disclosed herein may further generate feature clusters based on failed outputs and corresponding features and generate a number of performance views that illustrate performance of the machine learning system with respect to clustered groupings of the test dataset.

Characterizing failures of a machine learning model based on instance features

Disclosed are techniques for extracting, identifying, and consuming imprecise temporal elements ('ITEs'). A user input may be received from a client device. A prediction may be generated of one or more time intervals to which the user input refers based upon an ITE model. The user input may be associated with the prediction, and provided to the client device.

System for interpreting and managing imprecise temporal expressions

This document relates to automated generation of machine learning models, such as neural networks. One example system includes a hardware processing unit and a storage resource. The storage resource can store computer-readable instructions cause the hardware processing unit to perform an iterative model-growing process that involves modifying parent models to obtain child models. The iterative model-growing process can also include selecting candidate layers to include in the child models based at least on weights learned in an initialization process of the candidate layers. The system can also output a final model selected from the child models.

Automated generation of machine learning models

The subject disclosure is directed towards processing evidence, which may include high-dimensional streaming evidence, into a future belief state. The existing evidence is used to project a belief about a future state. The future belief state may be used to determine whether to wait for additional evidence, or to act now without waiting for additional evidence, e.g., based on a cost of the delay. For example, an autonomous assistant may decide based upon the belief whether to engage a person or not, or to wait for more information before the engagement decision is made.

Value of information with streaming evidence based on a prediction of a future belief at a future time

A server may be configured to receive first pose data from a virtual reality device, generate second pose data, including a pose prediction, based at least on the first pose data, render a map representation based at least on the pose prediction, determine regions of the map representation based on a likelihood of view of each of the regions, perform foveation in each of the regions and generate an encoded frame based on the regions and a network parameter, and send the encoded frame and second pose data to the device. The encoding may associate each of the regions with a quality level and base the foveation on each region's associated quality level. The foveation may use different quantization parameters for each of the regions and/or use rendering foveation for each of regions. Network conditions such as latency, bandwidth, and/or jitter may be considered for bit allocation in the foveation.

Virtual reality with interactive streaming video and likelihood-based foveation

The claimed subject matter provides systems and/or methods that facilitate inferring probability distributions over the destinations and/or routes of a user, from observations about context and partial trajectories of a trip. Destinations of a trip are based on at least one of a prior and a likelihood based at least in part on the received input data. The destination estimator component can use one or more of a personal destinations prior, time of day and day of week, a ground cover prior, driving efficiency associated with candidate locations, and a trip time likelihood to probabilistically predict the destination. In addition, data gathered from a population about the likelihood of visiting previously unvisited locations and the spatial configuration of such locations may be used to enhance the predictions of destinations and routes.

Methods for predicting destinations from partial trajectories employing open- and closed-world modeling methods

A computing system for calculating a total calorie count is provided. The computing system comprising an electronic personal assistant application program executed on a client computing device, and including a purchase tracking module configured to receive notifications of purchases made for individual food items, a calorie tracking module configured to identify calorie counts associated with each purchased food item and to calculate a calorie count statistic based thereon, a calorie apportionment module configured to apportion the calorie counts of the food items to a user so that the calorie count statistic is based upon an apportioned total of the calorie counts of the food items that have been apportioned to the user; and a graphical user interface module configured to display a graphical user interface including an indicator of the calorie count statistic on a display associated with the client computing device.

Calculating calorie statistics based on purchases

A client computing device is disclosed that comprises a processor and an electronic personal assistant application program. The personal assistant application program may be configured to capture user data associated with user activities across a plurality of computer programs. The user data may be sent to a personal assistant user data interpretation engine. A health-related suggestion based on at least a subset of the user data and anonymized statistics of a user population retrieved from an aggregated knowledge base may be received. The health-related suggestion may be displayed on a display associated with the client computing device.

Client computing device health-related suggestions

A computing system comprises an electronic personal assistant application program executed on a client computing device. Included are a health insurance retriever configured to retrieve health insurance information of a user, a history combiner configured to combined aggregated histories of a plurality of users into an anonymized combined time and location-based data, and a health recommender configured to identify at least one health condition of the user, determine a health recommendation at least on the identified health condition, and output the health recommendation including the recommended health care service, providers, and an indication of insurance coverage of the recommended service at each of the providers.

Recommendations for health benefit resources

Growing community of inventors

Kirkland, WA, United States of America

Eric J Horvitz