公开(公告)号：US20240160939A1

公开(公告)日：2024-05-16

申请号：US17987518

申请日：2022-11-15

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATISH KUMAR MOPUR ,  KRISHNAPRASAD LINGADAHALLI SHASTRY ,  SATHYANARAYANAN MANAMOHAN ,  RAVI SARVESWARA ,  GUNALAN PERUMAL VIJAYAN

IPC: G06N3/08

CPC classification number: G06N3/088

Abstract: Anomalies and drift detection in decentralized learning environments. The method includes deploying at a first node, (1) a local unsupervised autoencoder, trained at the first node, along with a local training data reference baseline for the first node, and (2) a global unsupervised autoencoder trained across a plurality of nodes, along with a corresponding global training data reference baseline. Production data at the first node is processed with local and global ML models deployed by a user. At least one of local and global anomaly data regarding anomalous production data or local and global drift data regarding drifting production data is derived based on the local and global training data reference baselines, respectively. At least one of the local anomaly data is compared with the global anomaly data or the local drift data with the global drift data for assessing impact of anomalies/drift on the ML models.

公开(公告)号：US20200050578A1

公开(公告)日：2020-02-13

申请号：US16100076

申请日：2018-08-09

Applicant: Hewlett Packard Enterprise Development LP

Inventor： SATISH KUMAR MOPUR ,  SAIKAT MUKHERJEE ,  GUNALAN PERUMAL VIJAYAN ,  SRIDHAR BALACHANDRIAH ,  ASHUTOSH AGRAWAL ,  KRISHNAPRASAD LINGADAHALLI SHASTRY ,  GREGORY S. BATTAS

IPC: G06F15/18 ,  G06F9/54 ,  H04L29/08

Abstract: The disclosure relates to technology that implements flow control for machine learning on data such as Internet of Things (“IoT”) datasets. The system may route outputs of a data splitter function performed on the IoT datasets to a designated target model based on a user specification for routing the outputs. In this manner, the IoT datasets may be dynamically routed to target datasets without reprogramming machine-learning pipelines, which enable rapid training, testing and validation of ML models as well as an ability to concurrently train, validate, and execute ML models.

公开(公告)号：US20160205189A1

公开(公告)日：2016-07-14

申请号：US14911719

申请日：2013-08-15

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATISH KUMAR MOPUR ,  Sumantha Kannantha ,  Shreyas Majithia ,  Akilesh Kailash ,  Aesha Dhar Roy ,  Satyaprakash Rao ,  Krishna Puttagunta ,  Chuan Peng ,  Prakash Hosahally Suryanarayana ,  Sudha Ramakrishnaiah ,  Ranganath Prabhu VV

IPC: H04L29/08 ,  H04L12/26 ,  G06F17/30

CPC classification number: H04L67/1097 ,  G06F11/30 ,  G06F11/3034 ,  G06F11/3041 ,  G06F11/3051 ,  G06F11/3058 ,  G06F11/3089 ,  G06F11/3452 ,  G06F11/3485 ,  G06F15/16 ,  G06F16/00 ,  G06F16/9024 ,  G06F2201/81 ,  H04L43/16

Abstract: The present subject matter relates to perform proactive monitoring and diagnostics in storage area networks (SANs). In one implementation, the method comprises depicting topology of the SAN in a graph, wherein the graph designates the devices as nodes, the connecting elements as edges, and depicts operations associated with at least one component of the nodes and edges. The method further comprises monitoring at least one parameter indicative of performance of the component to ascertain degradation of the at least one component and identifying, a hinge in the data associated with the monitoring, wherein the hinge is indicative of an initiation in degradation of the component. Based on the hinge, proactive diagnostics is preformed to compute a remaining lifetime of the at least one component. Thereafter, a notification is generated for an administrator of the SAN based on the remaining lifetime.

Abstract translation: 本主题涉及在存储区域网络（SAN）中执行主动监视和诊断。 在一个实现中，该方法包括描绘图中的SAN的拓扑，其中图表将设备指定为节点，连接元素作为边缘，并且描绘与节点和边缘的至少一个组件相关联的操作。 该方法还包括监测指示组件的性能的至少一个参数以确定至少一个组件的降解并且识别与监测相关联的数据中的铰链，其中所述铰链指示所述组件的降解起始 。 基于铰链，执行主动诊断以计算至少一个部件的剩余寿命。 此后，根据剩余生存时间为SAN的管理员生成通知。

公开(公告)号：US20230316710A1

公开(公告)日：2023-10-05

申请号：US17707612

申请日：2022-03-29

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATISH KUMAR MOPUR ,  Gunalan Perumal Vijayan ,  Shounak Bandopadhyay ,  Krishnaprasad Lingadahalli Shastry

IPC: G06V10/762 ,  G06V10/776 ,  G06V10/82 ,  G06N3/04

CPC classification number: G06V10/763 ,  G06V10/776 ,  G06V10/82 ,  G06N3/0454

Abstract: Systems and methods are provided for implementing a Siamese neural network using improved “sub” neural networks and loss function. For example, the system can detect a granular change in images using a Siamese Neural Network with Convolutional Autoencoders as the twin sub networks (e.g., Siamese AutoEncoder or “SAE”). In some examples, the loss function may be an adaptive loss function to the SAE network rather than a contrastive loss function, which can help enable smooth control of granularity of change detection across the images. In some examples, an image separation distance value may be calculated to determine the value of change between the image pairs. The image separation distance value may be determined using an Euclidean distance associated with a latent space of an encoder portion of the autoencoder of the neural networks.

公开(公告)号：US20230281958A1

公开(公告)日：2023-09-07

申请号：US17683816

申请日：2022-03-01

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATISH KUMAR MOPUR ,  Gunalan Perumal Vijayan ,  Krishna Prasad Lingadahalli Shastry

IPC: G06V10/762 ,  G06V10/82 ,  G06N7/00 ,  G06N3/08

CPC classification number: G06V10/762 ,  G06V10/82 ,  G06N7/005 ,  G06N3/08

Abstract: Systems and methods are provided for retraining machine learning (ML) models. Examples may automatically identify skewed, anomalous, and/or drift occurrence data in real-world input data. By automatically identifying such data, examples can reduce subjectivity in ML model retraining as well as reduce time spent determining a need to retrain a ML model. Accordingly, a determination can be made objectively by a computing system or device according to computer-implemented instructions. Additionally, examples may automatically isolate and transfer data relevant to the retraining of a ML model to a training environment for retraining the ML model using real-world input data. Examples also synthesize large samples of data for use in retraining a ML model. The synthesized data may be generated based on the isolated and transferred data and can be used in place of actual real-world input data to reduce a corresponding delay.

公开(公告)号：US20240354218A1

公开(公告)日：2024-10-24

申请号：US18302279

申请日：2023-04-18

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SRIDHAR BALACHANDRIAH ,  SATISH KUMAR MOPUR ,  LANCE MACKIMME EVANS ,  SHERIN THYIL GEORGE ,  KEVAN REHM

IPC: G06F11/34

CPC classification number: G06F11/3495 ,  G06F11/3433

Abstract: Systems and methods are provided for utilization of optimal data access interface usage in machine learning pipelines. Examples of the systems and methods disclosed herein include identifying data access interfaces comprising at least a first data access interface for a persistent storage distributed across a plurality of storage nodes and at least a second data access interface for an in-memory object store, and receiving, from a compute node, a data operation request as part of a machine learning pipeline. Additionally, performance metrics are obtained for the plurality of access interfaces, and based on a type of data operation request, the data operation is executed using a data access interface selected from the plurality of data access interface based on the performance metrics and providing an object handle to the compute node.

公开(公告)号：US20240312180A1

公开(公告)日：2024-09-19

申请号：US18184465

申请日：2023-03-15

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATISH KUMAR MOPUR ,  GUNALAN PERUMAL VIJAYAN ,  SHOUNAK BANDOPADHYAY ,  VIJAYA SHARVANI HINDNAVIS ,  KRISHNAPRASAD LINGADAHALLI SHASTRY

IPC: G06V10/762 ,  G06V10/26 ,  G06V10/42

CPC classification number: G06V10/762 ,  G06V10/26 ,  G06V10/42

Abstract: Systems and methods for preventing prediction performance degradation by detecting and extracting skews in data during both training and production environments is described herein. Feature extraction may be performed on training data during the training phase, followed by pattern analysis that assesses similarities across labeled training data sets. A reference pattern may be derived from the pattern analysis and feature extraction of the training data. Feature extraction and pattern analysis may be performed on production data during the serving phase, and a target pattern may be derived from the pattern analysis and feature extraction of the production data. The reference pattern and target pattern may be fed to a discrepancy detection functionality to detect discrepancies by using a sliding window to move the target pattern across the reference pattern to make comparisons between the patterns. The comparison may provide a quantitative skew across the training and production data.

公开(公告)号：US20240028417A1

公开(公告)日：2024-01-25

申请号：US17868587

申请日：2022-07-19

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： SATHYANARAYANAN MANAMOHAN ,  SATISH KUMAR MOPUR ,  KRISHNAPRASAD LINGADAHALLI SHASTRY ,  RAVI SARVESWARA

IPC: G06F9/50 ,  H04L9/00 ,  G06F9/455

CPC classification number: G06F9/5077 ,  H04L9/50 ,  G06F9/45558 ,  G06N20/20

Abstract: Systems and methods provide for a federated workflow solution to orchestrate entire machine learning (ML) workflows comprising multiple tasks, across silos. In other words, one or more sets/pluralities of tasks making up an ML workflow, can be executed across multiple resource partitions or domains. Federated workflow state can be maintained and shared through some form of distributed database/ledger, such as a blockchain. Agents that are locally deployed locally at the silos may orchestrate an ML workflow at a particular resource domains, each such agent having access, via the blockchain (acting as a globally visible/consistent state store), to the aforementioned workflow state. Such systems are capable of operating regardless of the existence of heterogeneous resources/aspects of a silo.