45 results about "Attribute learning" patented technology

Neural networks for optimal estimation (including prediction) and / or control involve an execution step and a learning step, and are characterized by the learning step being performed by neural computations. The set of learning rules cause the circuit's connection strengths to learn to approximate the optimal estimation and / or control function that minimizes estimation error and / or a measure of control cost. The classical Kalman filter and the classical Kalman optimal controller are important examples of such an optimal estimation and / or control function. The circuit uses only a stream of noisy measurements to infer relevant properties of the external dynamical system, learn the optimal estimation and / or control function, and apply its learning of this optimal function to input data streams in an online manner. In this way, the circuit simultaneously learns and generates estimates and / or control output signals that are optimal, given the network's current state of learning.

The invention relates to the field of data analysis, and provides an online learner activeness evaluation method based on a deep clustering algorithm, which comprises the following steps: acquiring multi-source information of learner attributes, online behaviors and online resources from an online platform by adopting database operation, code burying point and web crawler technologies; according to a full-view learning theory, the liveness of online learners is comprehensively evaluated from interaction behaviors between the learners and an online platform, learning contents and other learners, and liveness distribution of the learners in demographic statistics, regional affiliation and platform registration attributes is presented through a visualization method. And a new thought and a new method are provided for online platform application and service level evaluation.

The invention provides a picture processing method, an attribute learning method of an ornament mark, a terminal and a server. The method comprises the following steps that: when a receiving result shows that a picture to be processed is ornamented, selecting an operation instruction of the ornament mark, wherein the picture to be processed preloads an ornament mark; obtaining the attribution information of the ornament mark selected by the operation instruction, wherein the attribute information can be reusable or non-reusable; when the attribute information of the ornament mark is non-reusable, using the ornament mark selected by the operation instruction to replace the preloaded ornament mark on the picture to be processed; and when the attribute information of the ornament mark is reusable, loading the ornament mark selected by the operation instruction to the picture to be processed, wherein the picture to be processed preloads one ornament mark. In the embodiment of the invention, a user can load the same type of ornament mark on the picture to be processed without processing the preloaded ornament mark, user operation complexity is lowered, operation is simple, and user experience is improved.

The invention discloses a reservoir discontinuous boundary identification method based on an expansion convolutional neural network. The method comprises the following steps: obtaining an attribute graph from seismic data; mapping attribute data to a low-dimensional vector space by adopting a multi-layer fusion technology; the method comprises the following steps: analyzing geological data, logging data and seismic data to obtain a discontinuous boundary type, and obtaining a label according to the divided discontinuous boundary type; learning a deep feature r1 from an input attribute by adopting a CNN; learning a non-continuity feature r2 from an input attribute by using DCNN; carrying out splicing by adopting a splicing technology and characteristics; sending the splicing result to a pooling layer, and sending the splicing result to a full connection layer after average pooling; and outputting a result by using a Softmax function to obtain an identification type. The method has the advantages that features can be automatically learned, identification errors are reduced, and discontinuous boundary types are accurately distinguished; feature differences between boundary lines can be highlighted; and interference of false boundaries in seismic data can be reduced.

Learning and propagating visual attributes are disclosed. One embodiment of the present invention proposes a technique for performing spatial propagation. The technique includes generating a first directed acyclic graph (DAG) by connecting spatially adjacent points included in a set of unstructured points via directed edges in a first direction. The technique further includes applying a first set of neural network layers to one or more images associated with the set of unstructured points to generate (i) a set of features for the set of unstructured points and (ii) a set of pairs of relevancy between spatially adjacent points connected by directed edges. The technique further includes generating a set of tags for the set of unstructured points by propagating the set of features across the first DAG based on the set of pairs of relevancy.

The invention discloses a pedestrian attribute recognition system and method based on multi-layer feature learning. The system includes a bottom-up feature extraction module, a top-down feature fusion module, a feature prediction module, a multi-layer prediction fusion module and a test module. , the specific steps of this method are: process the image layer by layer from bottom to top to obtain multi-layer features; fuse the features of adjacent layers from top to bottom, compress the channel of the feature map obtained by the higher layer, and combine it with the upper layer The sampled feature map is subjected to feature fusion and channel dimensionality reduction, and the current layer features are output; the fused features and the extracted uppermost layer features are passed through the maximum pooling layer and the fully connected layer to obtain preliminary prediction results at different levels; The preliminary prediction results are superimposed, and each attribute of each layer of prediction is given a corresponding weight value to obtain the final prediction result; the prediction result corresponding to the picture is extracted, and the results of each index are calculated. The present invention learns a set of specific weights for each attribute based on the predicted value obtained by the fused features, so that each attribute can better use multi-layer features to obtain better recognition effect.

The invention provides a picture processing method, an attribute learning method of an ornament mark, a terminal and a server. The method comprises the following steps that: when a receiving result shows that a picture to be processed is ornamented, selecting an operation instruction of the ornament mark, wherein the picture to be processed preloads an ornament mark; obtaining the attribution information of the ornament mark selected by the operation instruction, wherein the attribute information can be reusable or non-reusable; when the attribute information of the ornament mark is non-reusable, using the ornament mark selected by the operation instruction to replace the preloaded ornament mark on the picture to be processed; and when the attribute information of the ornament mark is reusable, loading the ornament mark selected by the operation instruction to the picture to be processed, wherein the picture to be processed preloads one ornament mark. In the embodiment of the invention, a user can load the same type of ornament mark on the picture to be processed without processing the preloaded ornament mark, user operation complexity is lowered, operation is simple, and user experience is improved.