Verizon Patent and Licensing Inc.

Snowflake Inc.

Boyu Wang

Qiming Jiang

Boxin Jiang

John Reumann

Jiaxun Wu

Reza Fardid

Michel Adar

Nick G Suizo

Garrett Frere

Pradeep Dorairaj

Sujoy Bhattacharjee

Naveed Farooqui

Mengjia Xia

Carsten Stann

The subject technology receives first data. The subject technology transforms first data to a wide format, the wide format comprising a second table. The subject technology splits a set of rows into a set of shards of rows. The subject technology, for each shard from the set of shards, sends a particular user defined table function (UDTF), including a particular shard of rows, to a different execution node to perform a set of operations for determining a rolling correlation over a window size. The subject technology provides a set of output values of each particular UDTF corresponding to each shard from the set of shards in a second UDTF. The subject technology sends the second UDTF to a particular execution node to perform an aggregate operation. The subject technology receives a value of the rolling correlation from the particular execution node based on the aggregate operation.

Distributed in-database vectorized operations using user defined table functions

A system described herein may provide a technique for generating one or more predictive models of device availability, which may be used to predict whether a given device will be able to be reached via one or more networks to receive information, such as Over-the-Air ('OTA') updates. The predictive models may be based on, for example, radio frequency (“RF”) metrics, device availability metrics, and timing offsets between times associated with such RF metrics and availability metrics. For a given device, based on RF metrics associated with the device and further based on a candidate time, the predictive model may be used to determine whether the device will be available at the candidate time.

Systems and methods for over-the-air update scheduling of smart devices based on signal quality and device availability metrics

Using an attributes model of a time series forecasting model, determine a set of features based on time series data, the set of features including periodic components. The time series data may be divided into a set of segments. For each segment of the set of segments, a weight may be assigned using an age of the segment, resulting in a set of weighted segments of time series data. Using a trend detection model of the time series forecasting model, trend data from the set of weighted segments of time series data may be determined. A time series forecast may be generated by combining the set of features and the trend data.

Enhanced time series forecasting

A system described herein may generate one or more models based on relationship data associated with multiple objects. The models may include one or more thresholds (e.g., which may indicate potential collisions between objects). The system may compare relationship data, associated with a particular vehicle and a particular object, with the one or more thresholds associated with the one or more models, and determine whether the relationship data associated with the particular vehicle and the particular object is rare data with respect to the one or more models. When the relationship data associated with the particular vehicle and the particular object is not rare data, the system may refraining from causing the particular vehicle to perform a collision prevention measure, and when the relationship data associated with the particular vehicle and the particular object is rare data, the system may cause the particular vehicle to perform the collision prevention measure.

Systems and methods for vehicular collision detection based on multi-dimensional collision models

A processing device may analyze a set of time series data using a time series forecasting model comprising an attributes model and a trend detection model. The attributes model may comprise a modified gradient boosting decision tree (GBDT) based algorithm. Analyzing the set of time series data comprises determining a set of features of the set of time series data, the set of features including periodic components as well as arbitrary components. A trend of the set of time series data may be determined using the trend detection model and the set of features and the trend may be combined to generate a time series forecast.

A system described herein may provide a technique for generating one or more predictive models of device availability, which may be used to predict whether a given device will be able to be reached via one or more networks to receive information, such as Over-the-Air ('OTA') updates. The predictive models may be based on, for example, radio frequency ('RF') metrics, device availability metrics, and timing offsets between times associated with such RF metrics and availability metrics. For a given device, based on RF metrics associated with the device and further based on a candidate time, the predictive model may be used to determine whether the device will be available at the candidate time.

Growing community of inventors

Menlo Park, CA, United States of America

Boyu Wang