公开(公告)号：US20250112843A1

公开(公告)日：2025-04-03

申请号：US18476913

申请日：2023-09-28

Applicant: Microsoft Technology Licensing, LLC

Inventor： Sathiya Kumaran MANI ,  Tsuwang HSIEH ,  Ranveer CHANDRA ,  Srikanth KANDULA ,  Santiago Martin SEGARRA

IPC: H04L43/045 ,  H04L41/16 ,  H04L43/08 ,  H04L43/55

Abstract: Securing and optimizing communications for a cloud service provider includes collecting connection summary information at network interface devices associated with host computing devices for a group of resources allocated to a customer of the cloud computing environment. The connection summary information includes local address information, remote address information, and data information, each connection established via the network interface devices. At least one communication graph is generated for the group of resources using the connection summary information. The graph includes nodes that represent communication resources of the group of resources and edges extending between nodes that characterize communication between the nodes. At least one analytics process is performed on data from the graph to identify at least one of a micro-segmentation strategy, a communication pattern, and a flow prediction for the group of resources.

公开(公告)号：US20240096063A1

公开(公告)日：2024-03-21

申请号：US18078402

申请日：2022-12-09

Applicant: Microsoft Technology Licensing, LLC

Inventor： Ganesh ANANTHANARAYANAN ,  Yuanchao SHU ,  Paramvir BAHL ,  Tsuwang HSIEH

IPC: G06V10/77

CPC classification number: G06V10/7715 ,  G06V2201/10

Abstract: Systems and methods are provided for reusing and retraining an image recognition model for video analytics. The image recognition model is used for inferring a frame of video data that is captured at edge devices. The edge devices periodically or under predetermined conditions transmits a captured frame of video data to perform inferencing. The disclosed technology is directed to select an image recognition model from a model store for reusing or for retraining. A model selector uses a gating network model to determine ranked candidate models for validation. The validation includes iterations of retraining the image recognition model and stopping the iteration when a rate of improving accuracy by retraining becomes smaller than the previous iteration step. Retraining a model includes generating reference data using a teacher model and retraining the model using the reference data. Integrating reuse and retraining of models enables improvement in accuracy and efficiency.

公开(公告)号：US20250036375A1

公开(公告)日：2025-01-30

申请号：US18395321

申请日：2023-12-22

Applicant: Microsoft Technology Licensing, LLC

Inventor： Tsuwang HSIEH ,  Sathiya Kumaran MANI ,  Ranveer CHANDRA ,  Srikanth KANDULA ,  Santiago Martin SEGARRA ,  Yajie ZHOU

IPC: G06F8/35 ,  G06F16/901

Abstract: This patent relates to automating network management. One example includes a graph analysis and manipulation tool configured to receive a natural language prompt relating to a network management activity. The graph analysis and manipulation tool is also configured to access a graph resource and to generate code that addresses the network management activity as a graph manipulation task.

公开(公告)号：US20220414534A1

公开(公告)日：2022-12-29

申请号：US17362115

申请日：2021-06-29

Applicant: Microsoft Technology Licensing, LLC

Inventor： Ganesh ANANTHANARAYANAN ,  Yuanchao SHU ,  Paramvir BAHL ,  Tsuwang HSIEH

IPC: G06N20/00

Abstract: Systems and methods are provided for continuous learning of models across hierarchies under a multi-access edge computing. In particular, an on-premises edge server, using a model, generates inference data associated with captured stream data. A data drift determiner determines a data drift in the inference data by comparing the data against reference data generated using a golden model. The data drift indicates a loss of accuracy in the inference data. A gateway model maintains one or more models in a model cache for update the model. The gateway model instructs the one or more servers to train the new model. The gateway model transmits the trained model to update the model in the on-premises edge server. Training the new model includes determining an on-premises edge server with computing resources available to train the new model while generating other inference data for incoming stream data in the data analytic pipeline.

公开(公告)号：US20250088428A1

公开(公告)日：2025-03-13

申请号：US18367775

申请日：2023-09-13

Applicant: Microsoft Technology Licensing, LLC

Inventor： Tsuwang HSIEH ,  Santiago Martin SEGARRA ,  Sathiya Kumaran MANI ,  Srikanth KANDULA ,  Michael Dean WONG

IPC: H04L41/16 ,  H04L41/14 ,  H04L41/147 ,  H04L43/045

Abstract: This document relates to automating detecting anomalies in network behavior of an application Generally, the disclosed techniques can obtain network flow data for an application. A machine learning model can be used to process the network flow data to detect anomalies. The machine learning model can be retrained over time to adapt to changing network behavior of the application. In some cases, a graph neural network is employed to detect the anomalies.

公开(公告)号：US20220366300A1

公开(公告)日：2022-11-17

申请号：US17322184

申请日：2021-05-17

Applicant: Microsoft Technology Licensing, LLC

Inventor： Tsuwang HSIEH ,  Behnaz ARZANI ,  Ankur MALLICK

IPC: G06N20/00

Abstract: A cloud-based service uses an offline training pipeline to categorize training data for machine learning (ML) models into various clusters. Incoming test data that is received by a data center or in a cloud environment is compared against the categorized training data to identify the appropriate ML model to assign the test data. The comparison of the test data is done in real-time using a similarity metric that takes into account spatial and temporal factors of the test data relative to the categorized training data.