Google Inc.

Beijing Didi Infinity Technology and Development Co., Ltd.

Synopsys, Inc.

Applied Materials, Inc.

Didi Research America, LLC

Zoom Communications, Inc.

Tao Huang

Anshul Kothari

Anurag Agrawal

Seung Yi

Girish Baliga

Jian Chen

Yingwei Cui

Zhiwei Qin

Xiaodan Song

Fong Shen

Saurabh Mahajan

Ruslan Kudubayev

Leo Sei

Tao Lin

Min Pan

Yinhong Chang

Shuaiji Li

Fangfang Zhang

Zuo Dai

Jaehan Jeon

Rongjun Wang

Dien-Yeh Wu

Yu Lei

Yi Xu

Wei Lei

Deepak Chandra

Shirish A Pethe

Xinyu Tang

Aixi Zhang

Geraldine M Vasquez

Zhimin Qi

Charles Chapin

Xingyao Gao

Xianyuan Zhao

Yintai Ma

Tobias Bjerregaard

Konstantinos Tsirogiannis

Xiang Chang

Patrick Po-Chun Li

Hui Wen

Ran Yan

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for distributing content are disclosed. In one aspect, a method includes storing, in a data structure, data specifying a future live event. An opportunity to provide the specified content to a user at a user device is identified. It is determined that (i) a time of the opportunity is between a start time and an end time for the live event, (ii) that a user device is located in a same geographic region as the live event based on geographical data provided by the user device, and (iii) the user is interested in the live event based on attributes of the user matching attributes of other users that were identified as interested in the live event (e.g., based on evaluation of online search data). The content is provided for display at the user device.

Event-based content distribution

A system and method updates a clock tree based on skew values of the circuit design. The clock tree is updated by obtaining a circuit design that includes circuit elements and a clock tree. The clock tree includes clock sources and clock sinks. Data path slack values for the clock tree are determined based on the clock sources and the clock sinks. Further, clock arrival values for the clock tree are determined based on the clock sources and the clock sinks. A first total local skew value of the circuit design is determined based on the data path slack values, and the clock arrival values, and updating the clock tree based on the first total local skew value.

Circuit design having an improved clock tree

Methods for reducing resistivity of metal gapfill include depositing a conformal layer in an opening of a feature and on a field of a substrate with a first thickness of the conformal layer of approximately 10 microns or less, depositing a non-conformal metal layer directly on the conformal layer at a bottom of the opening and directly on the field using an anisotropic deposition process. A second thickness of the non-conformal metal layer on the field and on the bottom of the feature is approximately 30 microns or greater. And depositing a metal gapfill material in the opening of the feature and on the field where the metal gapfill material completely fills the opening without any voids.

Methods for forming metal gapfill with low resistivity

A system and method for changing a circuit design are described. The method includes generating a propagation graph for the circuit design and estimating slack values for some of the paths in the propagation graph. The method also includes making a virtual change to the circuit design and determining whether to accept or reject the change based on how the change affects the estimated slack values.

System for making circuit design changes

Systems and methods of parallel uniform resource identifier resource extraction are provided. A tracking uniform resource identifier (URI) generation component can generate a first uniform resource locator (URL) that, responsive to actuation by a client computing device, causes the client computing device to send a content item identifier of a content item and a device identifier of the client computing device, via a computer network, to a records server at a first network location identified by the first URL. A resource URI generation component can generate a second URL that, responsive to actuation by the client computing device, causes the client computing device to retrieve, via the computer network, a first resource from a second network location identified by the second URL for display by the client computing device. A content item packaging component can generate the content item.

Parallel execution of request tracking and resource delivery

A computer-implemented method for domain analysis comprises: obtaining, by a computing device, a domain; and inputting, by the computing device, the obtained domain to a trained detection model to determine if the obtained domain was generated by one or more domain generation algorithms. The detection model comprises a neural network model, a n-gram-based machine learning model, and an ensemble layer. Inputting the obtained domain to the detection model comprises inputting the obtained domain to each of the neural network model and the n-gram-based machine learning model. The neural network model and the n-gram-based machine learning model both output to the ensemble layer. The ensemble layer outputs a probability that the obtained domain was generated by the domain generation algorithms.

System and method for detecting generated domain

Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium for dynamic contact information assignment. A method includes: identifying a pool of telephone numbers; assigning the telephone numbers to a pool manager; allocating by the pool manager, subsets of the telephone numbers to a plurality of allocators, each allocator responsible for allocating telephone numbers to an associated group of content sponsors; determining a first allocation of a first subset, the first allocation being distributed among the content sponsors associated with a first allocator, creating first pools each associated with a respective one of the content sponsors associated with the first allocator; reclaiming one or more telephone numbers from a pool of the first pools; and assigning ones of the reclaimed telephone numbers by the first allocator into either other pools of the first pools or back to the pool manager for allocation to other allocators.

Automatic virtual phone number pool management

Systems and methods are provided for URI (Uniform Resource Identifier) consolidation. An exemplary method for URI consolidation may comprise receiving a URI comprising one or more URI segments in a string from left to right, parsing the URI to determine if any of the URI segments matches with any of a plurality of URI segment records in a URI database, and in response to determining one or more URI segments matching respectively with one or more of the URI segment records, using a portion of the received URI up to and including a rightmost URI segment matching with a URI segment record in the URI database to represent the received URI.

System and method for uniform resource identifier (URI) consolidation

Methods, systems, and apparatus include computer programs encoded on a computer-readable storage medium for dynamic telephone number assignment. A method includes: providing a content item to a user device, the content item including a control for initiating a call to a content sponsor associated with the content item; assigning a telephone number from a pool to the content item; receiving a control interaction indication; facilitating a connection between the user and the content sponsor based on the interaction; storing an association between a user identifier, the telephone number, and a content sponsor identifier such that future calls from the user to the telephone number can be routed to the content sponsor, even after the telephone number has been returned to the pool; and linking a content item impression with the connection such that a report can be generated that shows an effectiveness of the content item impression.

Dynamic telephone number assignment

A method for point-to-point traffic prediction comprises: obtaining, from a plurality of computing devices, time-series locations of a plurality of vehicles respectively associated with the computing devices, wherein: the time-series locations form first trajectory data comprising corresponding trajectories at least passing from a first point O to a second point D within a first time interval; obtaining a traffic volume between O and D for a second time interval that is temporally after the first time interval; training one or more weights of a neural network model by inputting the first trajectory data and the traffic volume to the neural network model and using the obtained traffic volume as ground truth to obtain a trained neural network model; and inputting second trajectory data between O and D to the trained neural network model to predict a future traffic volume between O and D for the a future time interval.

System and method for point-to-point traffic prediction

Growing community of inventors

San Jose, CA, United States of America

Tao Huang