Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing a machine learning task on a network input to generate a network output. One of the systems includes an attention neural network configured to perform the machine learning task. The attention neural network includes one or more attentions layers that each include a squared ReLU activation layer, a depth-wise convolution layer, or both.

Human interactive texture generation and search systems and methods are described. A deep convolutional generative adversarial network is used for mapping information in a latent space into texture models. An interactive evolutionary computation algorithm for searching a texture through an evolving latent space driven by human preference is also described. Advantages of a generative model and an evolutionary computation are combined to realize a controllable and bounded texture tuning process under the guidance of human preferences. Additionally, a fully haptic user interface is described, which can be used to evaluate the systems and methods in terms of their efficiency and accuracy of searching and generating new virtual textures that are closely representative of given real textures.

Methods, systems, and apparatus, including computer programs encoded on computer-storage media, for providing one or more values from a distribution of values to a neural network trained to generate simulated channel responses corresponding to one or more radio frequency (RF) communication channels; and obtaining an output of the neural network based on processing the one or more values by the neural network, the output indicating a simulated channel response corresponding to at least one communication channel of the one or more RF communication channels.

A method of forming a neural network includes specifying layers of neural network neurons. A parameter genome is defined with numerical parameters characterizing connections between neural network neurons in the layers of neural network neurons, where the connections are defined from a neuron in a current layer to neurons in a set of adjacent layers, and where the parameter genome has a unique representation characterized by kilobytes of numerical parameters. Parameter genomes are combined into a connectome characterizing all connections between all neural network neurons in the connectome, where the connectome has in excess of millions of neural network neurons and billions of connections between the neural network neurons.

Methods and systems that provide one or more recommended configurations to planners using large data sets in an efficient manner. These methods and systems provide optimization of objectives using a genetic algorithm that can provide parameter recommendations that optimize one or more objectives in an efficient and timely manner. The methods and systems disclosed herein are flexible enough to satisfy diverse use cases.

Systems, methods, and apparatuses for generating and using machine learning models using genetic data. A set of input features for training the machine learning model can be identified and used to train the model based on training samples, e.g., for which one or more labels are known. As examples, the input features can include aligned variables (e.g., derived from sequences aligned to a population level or individual references) and/or non-aligned variables (e.g., sequence content). The features can be classified into different groups based on the underlying genetic data or intermediate values resulting from a processing of the underlying genetic data. Features can be selected from a feature space for creating a feature vector for training a model. The selection and creation of feature vectors can be performed iteratively to train many models as part of a search for optimal features and an optimal model.

Disclosed in the present disclosure are a hybrid photovoltaic power prediction method and system based on multi-resource data fusion. The method includes: acquiring historical power sequence data and external meteorological data on a day to be predicted; inputting the data into a trained convolutional neural network prediction sub-model, long short-term memory network prediction sub-model and extreme gradient boosting tree prediction sub-model to predict photovoltaic power; classifying weather types according to a cloud cover on the day to be predicted, and determining prediction weights of the prediction sub-models; and fusing prediction results of the prediction sub-models based on the weights to obtain a final the prediction result of the photovoltaic power. The present disclosure integrates data of various different architectures, fully analyzes features of historical power data, meteorological data and satellite image data, and then fuses the data into unified data which is better and richer than single data.

A memory circuit configured to perform multiply-accumulate (MAC) operations for performance of an artificial neural network includes a series of synapse cells arranged in a cross-bar array. Each cell includes a memory transistor connected in series with a memristor. The memory circuit also includes input lines connected to the source terminal of the memory transistor in each cell, output lines connected to an output terminal of the memristor in each cell, and programming lines coupled to a gate terminal of the memory transistor in each cell. The memristor of each cell is configured to store a conductance value representative of a synaptic weight of a synapse connected to a neuron in the artificial neural network, and the memory transistor of each cell is configured to store a threshold voltage representative of a synaptic importance value of the synapse connected to the neuron in the artificial neural network.

The technology disclosed proposes a novel asynchronous evaluation strategy (AES) that increases throughput of evolutionary algorithms by continuously maintaining a queue of K individuals ready to be sent to the worker nodes for evaluation and evolving the next generation once a fraction Mi of the K individuals have been evaluated by the worker nodes, where Mi<<K. A suitable value for Mi is determined experimentally, balancing diversity and efficiency. The technology disclosed is extended to coevolution of deep neural network supermodules and blueprints in the form of AES for cooperative evolution of deep neural networks (CoDeepNEAT-AES). Applied to image captioning domain, a threefold speedup is observed on 200 graphics processing unit (GPU) worker nodes, demonstrating that the disclosed AES and CoDeepNEAT-AES are promising techniques for evolving complex systems with long and variable evaluation times.

Described is a method for predicting multiple components' content in a case that rare earth ions with and without color feature coexist, and relates to component content prediction in rare earth extraction process. It is difficult to quickly/accurately detect component's content in rare earth extraction process. Because of relatively large difference between images' color features of CePr/Nd mixed solution with colorless Ce ions and Pr/Nd solution, detecting content method of single rare earth element based on color feature is no longer applicable. The method includes: first searching for H and S components with maximum correlation with component content in HSI color space; establishing ELM based multi-component content soft measurement model using H and S component first-order moment as input; and for uncertainty of initial weight and ELM (extreme learning machine) model's threshold, optimizing model parameters using genetic algorithm GA to optimize ELM model for component content prediction higher precision.