A method for securing the functioning of an electronic device, which comprises an electronic board and one or more peripheral units connected to or integrated with the electronic board, an integrated storage unit being provided on the electronic board, in which a management program is stored which, when executed, manages, by means of a set of management instructions, the functioning of the electronic board and of the peripheral units.

Accelerated machine learning using an efficient preconditioner for Kernel Ridge Regression (KRR). A plurality of anchor points may be selected by: projecting an initial kernel onto a random matrix in a lower dimensional space to generate a randomized decomposition of the initial kernel, permuting the randomized decomposition to reorder its columns and/or rows to approximate the initial kernel, and selecting anchor points representing a subset of the columns and/or rows based on their permuted order. A reduced-rank approximation kernel may be generated comprising the subset of columns and/or rows represented by the selected anchor points. A KRR system may be preconditioned using a preconditioner generated based on the reduced-rank approximation kernel. The preconditioned KRR system may be solved to train the machine learning model. This KRR technique may be executed without generating the KRR kernel, reducing processor and memory consumption.

A diagnosis prediction (DP) computing device (102) receives training datasets from a health records server (108A), an insurance claims server (108B), and other third party servers (108C). DP computing device builds a model based on the training datasets and stores the model on a database (106) via a database server (104). Using the model and a stochastic learning algorithm, a risk estimator (110) determines a prediction of a disease or disorder diagnosis of a patient to a client device (112). The prediction is based on data gathered pertaining to the patient including unprocessed raw data comprising records of diagnostic codes generated during past medical encounters from an insurance claims database.

Techniques are disclosed for distributing data in a content delivery network configured to provide edge services using a plurality of service providers. Data indicative of data usage and cost data for the plurality of service providers is accessed. Based on the accessed data, an effective unit cost, multiplex efficiency, and channel utilization are determined for a selected user. A Bayesian optimization algorithm is applied to at least a portion of the accessed data. The content delivery network is configured to redistribute data traffic for the selected user based on a result of the applied Bayesian optimization algorithm.

A system and method for visualizing one or more crop response surfaces. The system includes a geospatial database associated with a crop prediction engine. The geospatial database receives soil composition information for plots of land. The crop prediction engine identifies covariates from the soil composition information, which has a feature matrix. The crop prediction engine generates a multi-dimensional covariate training data set from the covariates. The crop prediction engine then applies the multi-dimensional covariate training data set to a machine learning training model to generate at least one predictive crop-yield predictive model. The crop prediction engine ranks covariates having feature set interactions. Subsequently, the crop prediction engine determines a dominant crop-yield feature set interaction from the ranked covariates having feature set interactions. The crop prediction engine generates a crop response surface from the dominant crop-yield feature set interaction. The crop prediction engine then visualizes the crop response surface.

A system and method for visualizing one or more crop response surfaces. The system includes a geospatial database associated with a crop prediction engine. The geospatial database receives soil composition information for plots of land. The crop prediction engine identifies covariates from the soil composition information, which has a feature matrix. The crop prediction engine generates a multi-dimensional covariate training data set from the covariates. The crop prediction engine then applies the multi-dimensional covariate training data set to a machine learning training model to generate at least one predictive crop-yield predictive model. The crop prediction engine ranks covariates having feature set interactions. Subsequently, the crop prediction engine determines a dominant crop-yield feature set interaction from the ranked covariates having feature set interactions. The crop prediction engine generates a crop response surface from the dominant crop-yield feature set interaction. The crop prediction engine then visualizes the crop response surface.

Methods and systems are for improvements to in-vitro fertilization using morpho-kinetic signatures. These improvements are achieved by analyzing a series of images of a developing embryo (e.g., time-lapse images) as opposed to a single static image. For example, due to the difficulty in identifying clear distinctions between morphological states based on static images, as well as the unpredictability of morpho-kinetic development of an embryo, the system analyzes the development of an embryo as a whole over a given time frame (e.g., fertilization to blastulation), which provides a better prediction of the viability of a given embryo. The analysis may take the form of a morpho-kinetic signature, which itself may be used to determine an optimal time to transfer and/or implant an embryo into a patient.

Methods for generating data models using a generative adversarial network can begin by receiving a data model generation request by a model optimizer from an interface. The model optimizer can provision computing resources with a data model. As a further step, a synthetic dataset for training the data model can be generated using a generative network of a generative adversarial network, the generative network trained to generate output data differing at least a predetermined amount from a reference dataset according to a similarity metric. The computing resources can train the data model using the synthetic dataset. The model optimizer can evaluate performance criteria of the data model and, based on the evaluation of the performance criteria of the data model, store the data model and metadata of the data model in a model storage. The data model can then be used to process production data.

Methods for generating data models using a generative adversarial network can begin by receiving a data model generation request by a model optimizer from an interface. The model optimizer can provision computing resources with a data model. As a further step, a synthetic dataset for training the data model can be generated using a generative network of a generative adversarial network, the generative network trained to generate output data differing at least a predetermined amount from a reference dataset according to a similarity metric. The computing resources can train the data model using the synthetic dataset. The model optimizer can evaluate performance criteria of the data model and, based on the evaluation of the performance criteria of the data model, store the data model and metadata of the data model in a model storage. The data model can then be used to process production data.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining a terminal idle time. In some implementations, a server can obtain communication data from a plurality of devices in a communication network, wherein the communication data indicates levels of network traffic for the device over time. The server can generate an idle period forecasting model configured to predict occurrence of future communication idle periods in which communication activity is predicted to be below a threshold. The server can provide the idle period forecasting model to each of the plurality of devices such that the devices can respectively use the idle period forecasting model to locally predict future communication idle periods of the devices.