公开(公告)号：US20250071069A1

公开(公告)日：2025-02-27

申请号：US18483849

申请日：2023-10-10

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Aladin Djuhera ,  Alecio Pedro Delazari Binotto ,  Fernando Luiz Koch ,  Rob High

IPC: H04L47/70 ,  H04L41/16 ,  H04L47/80

Abstract: The present disclosure relates to a method comprising receiving a request to execute a workload using an artificial intelligence model. A current resource utilization status in the distributed system may be determined. The current resource utilization status may be used to define a deployment configuration of the artificial intelligence model, wherein the deployment configuration is defined by: a number and structure of input blocks, a number and structure of output blocks and the intermediate block of the artificial intelligence model, a second computer system to execute the intermediate block, and one or more first computer systems to execute the input and output blocks. The artificial intelligence model may be deployed in accordance with the defined deployment configuration and the workload may be executed.

公开(公告)号：US12238081B2

公开(公告)日：2025-02-25

申请号：US17457217

申请日：2021-12-01

Applicant: PayPal, Inc.

Inventor： Abhishek Chhibber ,  Darshankumar Bhadrasinh Desai ,  Michael Charles Todasco ,  Vidyut Mukund Naware ,  Nitin S. Sharma

IPC: H04L9/40 ,  H04L41/14 ,  H04L41/16 ,  H04L43/04 ,  H04L43/062 ,  H04L43/16

Abstract: Techniques are disclosed in which an edge user computing device pre-processes a stream of user data prior to using the stream of data to train a machine learning model at the edge device. The edge device receives the stream of user data, where the stream of data includes a first set of characteristics associated with the edge device and a second set of characteristics associated with a plurality of user requests received from a user of the edge device. The edge device repeatedly generates, using the stream of data, sets of pre-processed user data by performing pre-processing techniques on characteristics included in the stream of data. The edge device repeatedly trains, using the sets of pre-processed data, a baseline model to generate a device-trained model, where the baseline model is trained at the edge device without providing user data included in the stream of data to a server computer system.

公开(公告)号：US12238067B2

公开(公告)日：2025-02-25

申请号：US17548673

申请日：2021-12-13

Applicant: Perimeter 81 LTD

Inventor： Rony Pikarski

IPC: H04L9/40 ,  H04L41/16

Abstract: A computer implemented method of automatically generating security rules for a networked environment based on anomalies identified using Machine Learning (ML), comprising receiving one or more feature vectors each comprising a plurality of operational parameters of a plurality of objects of a networked environment, identifying one or more anomaly patterns in the networked environment by applying one or more trained ML models to the one or more feature vectors trained to identify patterns deviating from normal behavior of the plurality of objects, parsing each anomaly patterns to a set of behavioral rules by traversing the anomaly pattern through a tree-like decision model, and generating one or more security rules for the networked environment according to the set(s) of behavior rules. Wherein the one or more security rules are applied to increase security of the networked environment.

公开(公告)号：US20250062976A1

公开(公告)日：2025-02-20

申请号：US18808791

申请日：2024-08-19

Applicant: JPMORGAN CHASE BANK, N.A.

Inventor： Tyrone SAUNDERS ,  Allen SIEMENS ,  Christian Howard POST ,  Austin DAUZAT

IPC: H04L43/08 ,  H04L41/06 ,  H04L41/16

Abstract: A method may include: receiving a value stream for a process comprising an identification of systems involved in the process and expected events in the process; receiving an event from an event streaming platform published by a system of a plurality of systems and comprises a system identifier for the system and a process identifier for a process; processing the event by comparing the event to the value stream; generating a new event based on the processing; publishing the new event to the event streaming platform, wherein the new event is consumed by the plurality of systems; generating metrics for the process; publishing the metrics to an alerting platform, wherein the alerting platform is configured to generate an alert in response to receiving certain metrics; and publishing the metrics to an automation platform, wherein the automation platform is configured to suggest an automation in response to receiving certain metrics.

公开(公告)号：US20250062963A1

公开(公告)日：2025-02-20

申请号：US18938062

申请日：2024-11-05

Applicant: Red Hat, Inc.

Inventor： Marcel Hild

IPC: H04L41/12 ,  G06N3/08 ,  H04L41/16 ,  H04L43/04 ,  H04L43/18

Abstract: A method is provided that includes detecting, using metadata of a captured packet traffic, a communication pattern within each of one or more levels in a network stack by identifying a context among an identified packet involved in a network conversation, and detecting a communication pattern based on the network conversation, the context among the identified packet, and a payload of the packet. The method further includes generating, by a processing device, a topology of the network in view of the communication pattern detected within each of the one or more levels in the network stack.

公开(公告)号：US20250062958A1

公开(公告)日：2025-02-20

申请号：US18936507

申请日：2024-11-04

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： Sargam Jain ,  Charles Hogg ,  David Fehling, JR. ,  Bernd Bandemer ,  Jose Tellado

IPC: H04L41/082 ,  G06F8/65 ,  G06F11/00 ,  G06F11/34 ,  G06N20/10 ,  H04L41/0631 ,  H04L41/16

Abstract: Examples of the presently disclosed technology provide automated firmware recommendation systems that inject the intelligence of machine learning into the firmware recommendation process. To accomplish this, examples train a machine learning model on troves of historical customer firmware update data on a dynamic basis (e.g., examples may train the machine learning model on weekly basis to predict accepted firmware updates made by a vendor's customers across the most recent 6 months). From this dynamic training, the machine learning model can learn to predict/recommend an optimal firmware version for a customer/network device cluster based on firmware-related features, recent customer preferences, and other customer-specific factors. Once trained, examples can deploy the machine learning model to make highly tailored firmware recommendations for individual network device clusters of individual customers taking the above described factors into account.

公开(公告)号：US12231903B2

公开(公告)日：2025-02-18

申请号：US18783221

申请日：2024-07-24

Applicant: Digital Global Systems, Inc.

Inventor： Armando Montalvo

IPC: H04L12/28 ,  G06F30/27 ,  G06N3/02 ,  G06N5/022 ,  G06N5/04 ,  G06N20/00 ,  G06N20/10 ,  G06N20/20 ,  H04L41/0893 ,  H04L41/16 ,  H04W16/10 ,  H04W24/02 ,  H04W72/0453 ,  G06N3/042 ,  G06N3/045 ,  H04J1/16 ,  H04L41/0894 ,  H04W16/14 ,  H04W24/08

Abstract: Systems, methods, and apparatuses for providing dynamic, prioritized spectrum utilization management. The system includes at least one monitoring sensor, at least one data analysis engine, at least one application, a semantic engine, a programmable rules and policy editor, a tip and cue server, nd/or a control panel. The tip and cue server is operable utilize the environmental awareness from the data processed by the at least one data analysis engine in combination with additional information to create actionable data.

公开(公告)号：US12231300B2

公开(公告)日：2025-02-18

申请号：US18334735

申请日：2023-06-14

Applicant: Cisco Technology, Inc.

Inventor： Pengfei Sun ,  Qihong Shao ,  David C. White, Jr.

IPC: G06F15/16 ,  H04L41/12 ,  H04L41/14 ,  H04L41/16

Abstract: A heterogeneous graph learning system generates and analyzes network implementations. The heterogeneous graph learning system includes obtaining information describing multiple network implementations including heterogeneous nodes. The heterogeneous graph learning system also includes generating a one-hop graph connecting a particular node of the heterogeneous nodes with a set of related nodes. The one-hop graph connects the particular node with the set of related nodes via corresponding edges. The heterogeneous graph learning system further includes transforming the one-hop graph into a weighted graph based on a Dynamic Meta Path Transformation (DMPT). In the DMPT, each of the corresponding edges connecting the particular node to a corresponding related node among the set of related nodes is associated with a corresponding weight.

公开(公告)号：US12231292B2

公开(公告)日：2025-02-18

申请号：US18000856

申请日：2020-06-10

Applicant: Telefonaktiebolaget LM Ericsson (publ)

Inventor： Tamas Borsos ,  Zsófia Kallus ,  Péter Kersch ,  Peter Vaderna

IPC: H04L41/0823 ,  H04L41/0813 ,  H04L41/16 ,  H04L43/08

Abstract: There is provided a method for assessing network performance. The method includes acquiring network measurements indicative of any changes in a network following a change to a configuration of the network and data indicative of one or more factors capable of causing the changes in the network. The one or more factors are independent of the change to the configuration of the network. The method includes analysing the acquired network measurements and data to identify a contribution of the one or more factors to a key performance indicator (KPI) and a contribution of the change to the configuration of the network to the KPI. The KPI is predicted by a machine learning model and is a measure of the network performance following the change to the configuration of the network.

公开(公告)号：US20250055879A1

公开(公告)日：2025-02-13

申请号：US18366082

申请日：2023-08-07

Applicant: Dell Products, L.P.

Inventor： A Hema ,  Saumya Joshi

IPC: H04L9/40 ,  H04L41/16

Abstract: According to embodiments of the present disclosure, systems and methods for identifying, categorizing, and displaying links in an email message are disclosed. According to one embodiment, an Information Handling System (IHS) includes computer-executable instructions to receive a request to identify one or more links in an email message, identify one or more links in the email message, arrange each of the one or more links in one of a plurality of categories, and display the arranged links for view by a user.