International Business Machines Corporation

Amdocs Development Limited

Lior Turgeman

Nir Naaman

Nili Guy

Ira Rosen

Adar Amir

Idan Ben-Harrush

Michael Masin

Samuel Kallner

Yaniv Avrashi

Nili Ifergan-Guy

Mihir Parikh

Erez Gabay

Shmuel Kalner

An example system includes a processor to receive a data input and a predicted performance of a Deep Learning (DL) model deployed on a resource constrained edge device from a predictive model. The processor is to modify a control input for the DL model based on the data input and the predicted performance. The processor is to send the control input to the deployed DL model to modify performance of the DL model.

Controlling performance of deployed deep learning models on resource constrained edge device via predictive models

Techniques for generating a set of Deep Learning (DL) models are described. An example method includes training an initial set of DL models using the training data, wherein a topology of each of the DL models is determined based on the parameters vector. The method also includes generating a set of estimate performance functions for each of the DL models in the initial set based on the set of edge-related metrics, and generating a plurality of objective functions based on the set of estimated performance functions. The method also includes generating a final DL model set based on the objective functions, receiving a user selection of a selected DL model from the final DL model set, and deploying the selected DL model to an edge device.

Adaptation of deep learning models to resource constrained edge devices

An example system includes a processor to monitor system resources and performance preferences. The processor is to select model fragments based on the system resources and the performance preferences. The processor is to also construct a running artificial intelligence (AI) model from the selected model fragments. The processor is to further automatically modify the running AI model using the model fragments in response to detecting a change in the system resources or a change in the performance preferences.

Modifying artificial intelligence models using model fragments

As described herein, a system, method, and computer program are provided for dynamic prioritization of monitoring system related alerts. A plurality of alerts generated for a monitoring system are accessed. A first set of alert features predefined as high-level features are identified, wherein each of the high-level features is mapped to one or more alert features in a second set of alert features predefined as low-level features. The plurality of alerts are processed to determine a plurality of the most central high-level features. The plurality of alerts are grouped according to the plurality of the most central high-level features. Each group of alerts is processed to determine a plurality of the most central low-level features for the alerts in the group of alerts. A prioritized set of alerts are selected from the plurality of alerts based on the plurality of the most central low-level features.

System, method, and computer program for dynamic prioritization of monitoring system related alerts

A system for monitoring user engagement, comprising at least one hardware processor, configured to: receive from a plurality of sensors information about a sequence of interactions of a user with a software application during a time interval, the information comprising, for each of the sequence of interactions, a plurality of sensor output values of a plurality of engagement metrics reflecting the user's emotional attachment to the software application; compute a sequence of linearly uncorrelated principle components based on an outcome of principal component analysis (PCA) of the plurality of sensor output values of the plurality of engagement metrics; and select a first component from the sequence of linearly uncorrelated principle components to be a user engagement index representing engagement of the user with the application during the time interval according to the plurality of engagement metrics.

System and method for measuring user engagement

Growing community of inventors

Moreshet, Israel

Lior Turgeman