公开(公告)号：US12294529B2

公开(公告)日：2025-05-06

申请号：US18342516

申请日：2023-06-27

Applicant: ADOBE INC.

Inventor： Kanak Mahadik ,  Tong Yu ,  Junda Wu

IPC: H04L41/16 ,  H04L47/78

Abstract: Methods for determining optimal cloud service resource include determining a reward function for a set of resource configurations identifying cloud service resource parameters. The cloud service resource parameters include a source parameter and a target parameter of services to provide a client computing device. A source parameter dataset for the source parameter and a target parameter dataset is generated using the reward function and historical source parameter data. The matrices are then subject to SVD and clustering. A target parameter reward dataset is learned from output of the SVD and clustering. The target parameter dataset is used to determine the parameters for the target parameter for providing corresponding cloud service resources.

公开(公告)号：US12045701B2

公开(公告)日：2024-07-23

申请号：US17030730

申请日：2020-09-24

Applicant: Adobe Inc.

Inventor： Kanak Mahadik

IPC: G06N20/10 ,  G06F9/455

CPC classification number: G06N20/10 ,  G06F9/45558 ,  G06F2009/45587

Abstract: The disclosure describes one or more implementations of a serverless computing management system that utilizes an online learning model to dynamically adjust the number of serverless execution containers in a serverless pool based on incoming data patterns. For example, for each time instance in a given time period, the serverless computing management system utilizes the online learning model to balance computing latency and computing cost to determine how to intelligently resize the serverless pool, such that the online machine-learning models in the serverless pool can update in a manner that improves accuracy and computing efficiency while also minimizing unnecessary delays. Further, the serverless computing management system provides a framework that facilitates state-based training of online machine-learning models in a stateless and serverless cloud-based environment.

公开(公告)号：US20220035794A1

公开(公告)日：2022-02-03

申请号：US16943322

申请日：2020-07-30

Applicant: Adobe Inc.

Inventor： Kanak Mahadik

IPC: G06F16/23 ,  G06F16/901 ,  G06F16/903

Abstract: Certain embodiments involve tracking incremental updates to graph data structures and thereby facilitating efficient data retrieval. For instance, a computing system services a first query for one or more segments of computing devices, online entities, or both. The computing system services the first query by searching of a set of nodes from a graph data structure. The computing system receives a second query after the graph data structure has been modified. The computing system identifies, from a change list for tracking changes to the graph data structure, a subset of the nodes impacted by the modification to the graph data structure. The computing system services the second query by searching the subset of impacted nodes in the graph data structure.

公开(公告)号：US20240152769A1

公开(公告)日：2024-05-09

申请号：US18050607

申请日：2022-10-28

Applicant: ADOBE INC.

Inventor： Ryan A. Rossi ,  Kanak Mahadik ,  Mustafa Abdallah ElHosiny Abdallah ,  Sungchul Kim ,  Handong Zhao

IPC: G06N3/0985 ,  G06Q10/04

CPC classification number: G06N3/0985 ,  G06Q10/04

Abstract: Systems and methods for automatic forecasting are described. Embodiments of the present disclosure receive a time-series dataset; compute a time-series meta-feature vector based on the time-series dataset; generate a performance score for a forecasting model using a meta-learner machine learning model that takes the time-series meta-feature vector as input; select the forecasting model from a plurality of forecasting models based on the performance score; and generate predicted time-series data based on the time-series dataset using the selected forecasting model.

公开(公告)号：US11561965B2

公开(公告)日：2023-01-24

申请号：US16943322

申请日：2020-07-30

Applicant: Adobe Inc.

Inventor： Kanak Mahadik

IPC: G06F16/23 ,  G06F16/903 ,  G06F16/901

Abstract: Certain embodiments involve tracking incremental updates to graph data structures and thereby facilitating efficient data retrieval. For instance, a computing system services a first query for one or more segments of computing devices, online entities, or both. The computing system services the first query by searching of a set of nodes from a graph data structure. The computing system receives a second query after the graph data structure has been modified. The computing system identifies, from a change list for tracking changes to the graph data structure, a subset of the nodes impacted by the modification to the graph data structure. The computing system services the second query by searching the subset of impacted nodes in the graph data structure.

公开(公告)号：US20220092480A1

公开(公告)日：2022-03-24

申请号：US17030730

申请日：2020-09-24

Applicant: Adobe Inc.

Inventor： Kanak Mahadik

IPC: G06N20/10 ,  G06F9/455

Abstract: The disclosure describes one or more implementations of a serverless computing management system that utilizes an online learning model to dynamically adjust the number of serverless execution containers in a serverless pool based on incoming data patterns. For example, for each time instance in a given time period, the serverless computing management system utilizes the online learning model to balance computing latency and computing cost to determine how to intelligently resize the serverless pool, such that the online machine-learning models in the serverless pool can update in a manner that improves accuracy and computing efficiency while also minimizing unnecessary delays. Further, the serverless computing management system provides a framework that facilitates state-based training of online machine-learning models in a stateless and serverless cloud-based environment.