Hewlett Packard Enterprise Development LP

C3.ai, Inc.

Other

Hewlett-packard Development Company, L.P.

Micro Focus Lcc

Gowtham Bellala

Martin Arlitt

Cullen Edwin Bash

Manish Marwah

Geoff M Lyon

Shagufta Mehnaz

Chandrakant D Patel

Amip J Shah

Henrik Ohlsson

Nikhil Krishnan

Sina Khoshfetrat Pakazad

Dibyajyoti Banerjee

Bernardo A Huberman

Jerome Alexander Rolia

Jung Gun Lee

Wei Lu

Wei-Nchih Lee

Mi Zhang

Jose Alberto Cueto Barcenas

Sherif Abdelwahab

The present disclosure provides systems and methods that may advantageously apply machine learning to accurately manage and predict inventory variables with future uncertainty. In an aspect, the present disclosure provides a system that can receive an inventory dataset comprising a plurality of inventory variables that indicate at least historical (i) inventory levels, (ii) inventory holding costs, (iii) supplier orders, and/or (iv) lead times overtime. The plurality of inventory variables can be characterized by having one or more future uncertainty levels. The system can process the inventory dataset using a trained machine learning model to generate a prediction of the plurality inventory variables. The system can provide the processed in inventory dataset to an optimization algorithm. The optimization algorithm can be used to predict a target inventory level for optimizing an inventory holding cost. The optimization algorithm can comprise one or more constraint conditions.

Systems and methods for inventory management and optimization

The present disclosure provides systems and methods that may advantageously apply machine learning to accurately manage and predict inventory variables with future uncertainty. In an aspect, the present disclosure provides a system that can receive an inventory dataset comprising a plurality of inventory variables that indicate at least historical (i) inventory levels, (ii) inventory holding costs, (iii) supplier orders, and/or (iv) lead times over time. The plurality of inventory variables can be characterized by having one or more future uncertainty levels. The system can process the inventory dataset using a trained machine learning model to generate a prediction of the plurality inventory variables. The system can provide the processed inventory dataset to an optimization algorithm. The optimization algorithm can be used to predict a target inventory level for optimizing an inventory holding cost. The optimization algorithm can comprise one or more constraint conditions.

Network flow classification can include clustering a network flow database into a number of at least one of applications and network flows. Network flow classification can include classifying the number of the at least one of applications and network flows.

Network flow classification

Performance testing based on variable length segmentation and clustering of time series data is disclosed. One example is a system including a training module, a performance testing module, and an interface module. The training module generates a trained model to learn characteristics of a system of interest from training time series data by segmenting the training time series data into homogeneous windows of variable length, clustering the segments to identify patterns, and associating each cluster with a cluster score. The performance testing module analyzes system characteristics from testing time series data by receiving the testing time series data, and determining a performance metric for the testing time series data by analyzing the testing time series data based on the trained model. The interface module is communicatively linked to the performance testing module, and provides the performance metric via an interactive graphical user interface.

Performance testing based on variable length segmentation and clustering of time series data

Examples disclosed herein relate to: computing, by a computing device at a party among a plurality of parties, a sum of local data owned by the party. The local data is horizontally partitioned into a plurality of data segments, with each data segment representing a non-overlapping subset of data entries owned by a particular party; computing a local gradient based on the horizontally partitioned local data; initializing each data segment; anonymizing aggregated local gradients received from the mediator, wherein the aggregated local gradients comprise gradients computed based on a plurality of data entries owned by the plurality of parties; receiving, from a mediator, a global gradient based on the aggregated local gradients; learning a global analytic model based on the global gradient; and performing privacy-preserving multi-party analytics on the horizontally partitioned local data based on the learned global analytic model.

Performing privacy-preserving multi-party analytics on horizontally partitioned local data

The present disclosure discloses a method comprising: dividing, by a computing device at a first party among a plurality of parties, local data into a plurality of data segments; recursively encrypting, by the computing device, each data segment using a plurality of public keys corresponding to the plurality of parties and a mediator; sharing, by the computing device, the local data comprising the encrypted plurality of data segments with the mediator; anonymizing, by the computing device, aggregated local data received from the mediator; and communicating, by the computing device from the mediator, a global sum that preserves privacy of the plurality of parties in a multi-party environment, wherein the global sum is computed by the mediator based on the collection of data segments that are decrypted recursively using the private key corresponding to each party and the private key corresponding to the mediator.

Computing a global sum that preserves privacy of parties in a multi-party environment

Example computing devices described herein enable computation of a machine learning model on distributed multi-party data that is vertically partitioned, in a privacy preserving fashion. The computing device computes at a party a sum of local data owned by the party, wherein the local data is vertically partitioned into a plurality of data segments, each data segment representing a non-overlapping subset of data features; transforms a cost function of a data analytics task to a gradient descent function, wherein the cost function comprises a summation of a plurality of cost function values; anonymizes aggregated data shards received from a mediator; updating local model parameters based on the aggregated data shards; and performs privacy-preserving multi-party analytics on the vertically partitioned local data based on a learned global analytic model. It leverages a secure-sum protocol that provides strong security guarantees against collusion and prior-knowledge attacks.

Performing privacy-preserving multi-party analytics on vertically partitioned local data

Techniques for secure linking of attributes between a first node and a plurality of nodes are provided. In one aspect, the first node and the plurality of nodes maintain a distributed data set. The techniques may include encoding the attributes as integer values at the first node. The nodes of the plurality of nodes that include matching attributes may be determined using a secure list matching protocol.

Attribute determination using secure list matching protocol

Overlay networks of application components are managed. Applicant components may create overlay networks based on policies of the application components and an environment of the overlay network. The overlay network may be adjusted based on changes to the policies or the environment.

Overlay network management

Determining sensor placement in distributed networks includes identifying measurement locations in a distributed network with a tree topology and selecting a subset of the measurement locations for sensor placement with a mutual information function.

Determining sensor placement in distributed networks

A method for detecting power consumption anomalies. The method obtains two different time power series and compares them to generate a dissimilarity matrix. A dimensionality reduction equation is applied to the dissimilarity matrix to obtain a low dimensional embedding of the power data. A probability score of each of the time series being anomalous is then calculated.

Detecting anomalies in power consumption of electrical systems

In an implementation, a method for managing an entity may include establishing of a framework for characterization of the entity, in which the framework abstracts operation of the entity to a state machine. In addition, each state in the state machine may be characterized in terms of distributions of a performance metric associated with the entity. The method may also include implementing of the framework for characterization of the entity to manage the entity.

Growing community of inventors

Palo Alto, CA, United States of America

Gowtham Bellala