-
公开(公告)号：US20230066759A1
公开(公告)日：2023-03-02
申请号：US17463738
申请日：2021-09-01
Applicant: Cisco Technology, Inc.
Inventor： Nassim Benoussaid , David John Zacks , Zizhen Gao , Carlos M. Pignataro , Dmitry Goloubev
IPC: G06N20/00
Abstract: Techniques are provided for segmentation of data points after a dimension reduction. A proxy model is then trained based on results of the segmentation. The proxy model provides low latency high throughput labeling of additional data points, without the need to reduce dimensions of the additional data points. A second segmentation is performed with results of the second segmentation compared to that of the first segmentation. When results of the comparison meet certain criterion, configuration parameters of the segmentation are modified. For example, in some embodiments, a user interface is provided that displays shapley values indicating a mapping from the high dimension data to the segmented data. Input is then received that modifies the configuration parameters.
-
Patent Agency Ranking
公开(公告)号：US20230198946A1
公开(公告)日：2023-06-22
申请号：US17557865
申请日：2021-12-21
Applicant: Cisco Technology, Inc.
Inventor： David John Zacks , Dmitry Goloubev , Zizhen Gao , Carlos M. Pignataro , Nagendra Kumar Nainar
IPC: H04L9/40 , H04L47/2441 , H04L47/2483
CPC classification number: H04L63/0236 , H04L63/20 , H04L47/2441 , H04L47/2483 , H04L63/0245
Abstract: Methods are provided for predictive policy enforcement using encapsulated metadata. The methods involve obtaining a packet of an encapsulated traffic flow that is transported in a software-defined wide area network (SD-WAN) or in a cloud network. The packet includes a network virtualization tunneling header with an appended service plane protocol header and a payload. The methods further involve extracting, from the appended service plane protocol header, without performing deep packet inspection, enriched metadata that includes fields for one or more attributes related to a source of the packet or a destination of the packet, determining at least one network policy based on the enriched metadata, and applying, to the packet, the at least one network policy that relates to gathering analytics and/or transporting the encapsulated traffic flow to the destination.
-
公开(公告)号：US11722359B2
公开(公告)日：2023-08-08
申请号：US17479297
申请日：2021-09-20
Applicant: Cisco Technology, Inc.
Inventor： Enzo Fenoglio , David John Zacks , Zizhen Gao , Carlos M. Pignataro , Dmitry Goloubev
IPC: H04L29/08 , H04L41/0631 , H04L43/04 , H04L41/16 , G06F18/214
CPC classification number: H04L41/064 , G06F18/214 , H04L41/16 , H04L43/04
Abstract: A method, computer system, and computer program product are provided for detecting drift in predictive models for network devices and traffic. A plurality of streams of time-series telemetry data are obtained, the time-series telemetry data generated by network devices of a data network. The plurality of streams are analyzed to identify a subset of streams, wherein each stream of the subset of streams includes telemetry data that is substantially empirically distributed. The subset of streams of time-series data are analyzed to identify a change point. In response to identifying the change point, additional time-series data is obtained from one or more streams of the plurality of streams of time-series telemetry data. A predictive model is trained using the additional time-series data to update the predictive model and provide a trained predictive model.
-
公开(公告)号：US20230099153A1
公开(公告)日：2023-03-30
申请号：US17490349
申请日：2021-09-30
Applicant: Cisco Technology, Inc.
Inventor： Donald Mark Allen , Dmitry Goloubev
Abstract: Methods are provided in which a computing device obtains telemetry data associated with an enterprise network that includes a plurality of assets involved in providing one or more enterprise services, obtains available software upgrade information, and generates at least two remediation plans based on the telemetry data and the available software upgrade information. Each of the at least two remediation plans being directed to a change in a configuration of one or more assets of the plurality of assets. The methods further include computing a probability of success of each of the at least two remediation plans based on the telemetry data and the available software upgrade information and providing the at least two remediation plans with a respective probability of success.
-
公开(公告)号：US20230093130A1
公开(公告)日：2023-03-23
申请号：US17479297
申请日：2021-09-20
Applicant: Cisco Technology, Inc.
Inventor： Enzo Fenoglio , David John Zacks , Zizhen Gao , Carlos M. Pignataro , Dmitry Goloubev
Abstract: A method, computer system, and computer program product are provided for detecting drift in predictive models for network devices and traffic. A plurality of streams of time-series telemetry data are obtained, the time-series telemetry data generated by network devices of a data network. The plurality of streams are analyzed to identify a subset of streams, wherein each stream of the subset of streams includes telemetry data that is substantially empirically distributed. The subset of streams of time-series data are analyzed to identify a change point. In response to identifying the change point, additional time-series data is obtained from one or more streams of the plurality of streams of time-series telemetry data. A predictive model is trained using the additional time-series data to update the predictive model and provide a trained predictive model.
-
公开(公告)号：US20230092777A1
公开(公告)日：2023-03-23
申请号：US17479237
申请日：2021-09-20
Applicant: Cisco Technology, Inc.
Inventor： Nagendra Kumar Nainar , Carlos M. Pignataro , David John Zacks , Dmitry Goloubev
Abstract: A method, computer system, and computer program product are provided for decentralized machine learning. A plurality of computing networks are identified by determining that each computing network of the plurality of computing networks satisfies a predetermined number of criteria. A decentralized learning agent is provided to each computing network, wherein the decentralized learning agent is provided with input parameters for training and is trained using training data associated with a computing network to which the decentralized learning agent is provided. A plurality of learned parameters are obtained from the plurality of computing networks, wherein each learned parameter of the plurality of learned parameters is obtained by training the decentralized learning agent provided to each respective computing network. A global model is generated based on the plurality of learned parameters.
-
-
-
-
-
Search Results
6
records
(took 0.012 seconds)
