A method for visualizing microbiome data is described. Respective microbes and/or genes in microbiome data stored In in a database are identified. A network comprising nodes interconnected by edges is generated in a memory of a computer, each node representing one or more identified microbes or one or more microbial metabolites, and each edge of the network representing an association between a respective pair of the one or more identified microbes or a reaction mediated between two metabolites by an enzyme encoded in the one or more identified genes, with at least some nodes and edges of the network being each associated with a condition attribute identifying a groups and/or a timestamp associated with a sample in the database. The displayed network is dynamically updated in accordance with a filtering of the microbiome data based on the condition attributed and/or the timestamp attributed. Corresponding systems and computer-readable storages are also described.

A molecular structure acquisition method, an electronic device and a storage medium, which relate to the field of artificial intelligence such as deep learning, are disclosed. The method may include: performing, for an initial seed, the following first processing: generating M molecular structures according to the seed, M being a positive integer greater than one; taking the M molecular structures as candidate molecular structures, and selecting some molecular structures from the candidate molecular structures as progeny molecular structures; and performing evolutionary learning on the progeny molecular structures, taking the progeny molecular structures after evolutionary learning as the seed, and repeating the first processing until convergence reaches an optimization objective, and when the convergence reaches the optimization objective, a newly selected molecular structure is taken as a desired molecular structure.

A computer-implemented method to establish a relative importance of an input parameter p.sub.j in a plurality of input parameters p.sub.i in a data set input to a machine learning model, the data set represented by a j row by k column matrix I.sub.m, an intersection of each row with each column defining an element, the method includes for each of the plurality of parameters p.sub.i in the input data set, a computer sorts columns k.sub.i of the matrix I.sub.m. to produce a re-ordered matrix I.sub.m,j; the computer determines a hyper-parameter N* of sub-matrices into which may be sorted the values in a j.sup.th row of the re-ordered matrix I.sub.m,j; the computer generates a plurality of group sub-matrices G.sub.i, each of the group sub-matrices comprising a subset of columns and the jth row; the computer inputs the re-ordered matrix I.sub.m,j into a fully-trained machine learning model to produce machine learning model outputs; and the computer produces normalized mean values of the outputs.

Embodiments for implementing mixed precision learning for neural networks by a processor. A neural network may be replicated into a plurality of replicated instances and each of the plurality of replicated instances differ in precision used for representing and determining parameters of the neural network. Data instances may be routed to one or more of the plurality of replicated instances for processing according to a data pre-processing operation.

An example computer-implemented method includes receiving a plurality of agricultural data records including yield properties of products grown in fields and raw field features of the fields. The method also includes transforming the raw field features into distinct feature classes that characterize key features affecting yield of the one or more products, and generating, using data from the plurality of agricultural data records and the distinct feature classes, genomic-by-environmental relationships between one or more products, yield properties of the one or more products, and field features associated with the one or more products. Further, the method includes generating, based at least in part on the genomic-by-environmental relationships, predicted yield performance for a set of products associated with one or more target environments, generating product recommendations for the one or more target environments based on the predicted yield performance for the set of products, and providing one or more instructions configured to cause display of the product recommendations.

The technology described herein is directed towards optimizing power consumption of devices, e.g., in a datacenter. A modified (two-tier) genetic algorithm performs a carbon footprint-based optimization in a first tier to determine a candidate range of coefficients for each device type, e.g., servers, switches and storage devices/systems that likely reduce carbon footprint of each device type. In a second tier of the genetic algorithm, those ranges of coefficients are used in conjunction with actual power usage-based carbon footprint scores of individual devices to find respective sets of coefficients that minimize respective objective functions for the servers, the switches and the storage devices. The sets of coefficients can be used for power capping the devices. Device performance constraint-based intelligent selection can be used in one or both tiers to speed up convergence.

The technology described herein is directed towards optimizing power consumption of devices, e.g., in a datacenter. A modified (two-tier) genetic algorithm performs a carbon footprint-based optimization in a first tier to determine a candidate range of coefficients for each device type, e.g., servers, switches and storage devices/systems that likely reduce carbon footprint of each device type. In a second tier of the genetic algorithm, those ranges of coefficients are used in conjunction with actual power usage-based carbon footprint scores of individual devices to find respective sets of coefficients that minimize respective objective functions for the servers, the switches and the storage devices. The sets of coefficients can be used for power capping the devices. Device performance constraint-based intelligent selection can be used in one or both tiers to speed up convergence.

A parallel analog circuit automatic optimization method based on genetic algorithm and machine learning comprises global optimization based on genetic algorithm and local optimization based on machine learning, with the global optimization and the local optimization performed alternately. The global optimization based on genetic algorithm utilizes parallel SPICE simulations to improve the optimization efficiency while guaranteeing the optimization accuracy, combined with parallel computing. The local optimization based on machine learning establishes a machine learning model near the global optimal point obtained by the global optimization, and uses the machine learning model to replace the SPICE simulator, thus reducing the time costs brought by a large number of simulations.

A parallel analog circuit automatic optimization method based on genetic algorithm and machine learning comprises global optimization based on genetic algorithm and local optimization based on machine learning, with the global optimization and the local optimization performed alternately. The global optimization based on genetic algorithm utilizes parallel SPICE simulations to improve the optimization efficiency while guaranteeing the optimization accuracy, combined with parallel computing. The local optimization based on machine learning establishes a machine learning model near the global optimal point obtained by the global optimization, and uses the machine learning model to replace the SPICE simulator, thus reducing the time costs brought by a large number of simulations.

Techniques for recommending side-by-side plantings of pairs of hybrids or seeds include a server computer receiving agricultural data records that represent crop seed data describing seed and yield properties of hybrid seeds and first data for agricultural fields where the hybrid seeds were planted. The server receives second data for available hybrids and seeds and automatically calculates a dataset of success probability scores that describe the probability of a successful yield on the target fields. Data is organized as pairs to facilitate comparison of actual plantings to optimized plantings that have a probability of success (POS), in terms of yield lift or increased yield season-over-season, for different yield values. Confidence values are generated and stored in association with the POS values and can be used as a basis of visual output to support planting and/or field management decisions as part of an automated intelligent agricultural decision support system.