Abstract: The present application provides a model training, image processing method, device, storage medium, and program product relating to deep learning technology, which are able to screen auxiliary image data with image data for learning a target task, and further fuse the target image data and the auxiliary image data, so as to train a built and to-be-trained model with the fusion-processed fused image data. This implementation can increase the amount of data for training the model, and the data for training the model is determined is based on the target image data, which is suitable for learning the target task. Therefore, the solution provided by the present application can train an accurate target model even if the amount of target image data is not sufficient.

Abstract: A convertible aircraft system is provided that can convert to a helicopter configuration, an airplane configuration, or a gyroplane configuration before, during, or after flight. The convertible aircraft system includes a fuselage, a proximal flight assembly, a distal flight assembly, a support spar, and a tail assembly. The fuselage is the main structural body of the present invention. The proximal flight assembly and the distal flight assembly are the flight system of the present invention. The support spar provides an axis of rotation and a pole support for the proximal flight assembly and the distal flight assembly. The tail assembly provides stability during flight of the present invention. In more detail, the tail assembly may comprise at least one vertical stabilizer, at least one horizontal stabilizer, and at least one rudder in order to provide stability during flight of the present invention.

Abstract: A method, apparatus, device and readable medium for transfer learning in machine learning are provided. The method includes: constructing a target model according to the number of classes to be achieved by a target task and a duly-trained source model; obtaining a value of a regularized loss function of the corresponding target model and a value of a cross-entropy loss function of the target model, based on sets of training data in a training dataset of the target task; according to the value of the regularized loss function and the value of the cross-entropy loss function corresponding to each set of training data, updating parameters in the target model by a gradient descent method to implement the training of the target model. The above technical solution avoids excessive constraints on parameters in the prior art, thereby refraining from damaging the training effect of the source model on the target task.

Abstract: Provided are an image occlusion method, a model training method, a device, and a storage medium, which relate to the technical field of artificial intelligence, in particular, to the field of computer vision technologies and deep learning, and may be applied to image recognition, model training and other scenarios. The specific implementation solution is as follows: generating a candidate occlusion region according to an occlusion parameter; according to the candidate occlusion region, occluding an image to be processed to obtain a candidate occlusion image; determining a target occlusion region from the candidate occlusion region according to visual security and data availability of the candidate occlusion image; and according to the target occlusion region, occluding the image to be processed to obtain a target occlusion image. In this manner, the image to be processed is desensitized while the accuracy of target recognition is ensured.

Abstract: An instant no-sew and removable fabric fastener is similar to a spring push-pin fabric fastener, and thus, the instant no-sew and removable fabric fastener includes a fastener socket and a fastener stud. However, the instant no-sew and removable fabric fastener is different than a standard spring push-pin fabric fastener by incorporating a rotational-type attachment mechanism and by being completely removable. The rotational-type attachment mechanism allows the fastener stud and the fastener socket to be deadlocked to each other. The fastener stud cannot be detached from the fastener socket even with a force until either the fastener socket or the fastener stud is rotated 90 degrees. After the fastener socket or the fastener is rotated 90 degrees, the fastener stud can be detached from the fastener socket. Additionally, this means the instant no-sew and removable fabric fastener does not permanently attach to fabric and thus, can be removed when desired.

Abstract: A method for processing synthetic features is provided, and includes: the synthetic features to be evaluated and original features corresponding to the synthetic features are obtained. A feature extraction is performed on the synthetic features to be evaluated based on a number S of pre-trained samples, to obtain meta features with S samples. S is a positive integer. The meta features are input into the pre-trained meta feature evaluation model for a binary classification prediction, to obtain a probability of binary classification. Quality screening is performed on the synthetic features to be evaluated according to the probability of the binary classification, to obtain second synthetic features to be evaluated. The second synthetic features are classified in a good category. The second synthetic features and original features are input into a first classifier for evaluation. classified in a poor category.

Abstract: A method for transfer learning includes: obtaining a pre-trained model, and generating a model to be transferred based on the pre-trained model, in which the model to be transferred includes N Transformer layers, and N is a positive integer; obtaining a mini-batch by performing random sampling on a target training set; and training the model to be transferred based on the mini-batch, in which a loss value for each Transformer layer is generated based on an empirical loss value and a noise stability loss value.

Abstract: A method and an apparatus for training a classification model and data classification includes: obtaining a sample set and a pre-trained classification model, wherein the classification model includes at least two convolutional layers, each convolutional layer is connected to a classification layer through a fully connected layer; inputting the sample set into the classification model, and obtaining a prediction result output by each classification layer, wherein the prediction result includes a prediction probability of a class to which each sample belongs; calculating a probability threshold of each classification layer based on the prediction result output by each classification layer; setting a prediction stopping condition for the classification mode according to the probability threshold of each classification layer.

Abstract: A convertible aircraft system is provided that can convert to a helicopter configuration, an airplane configuration, or a gyroplane configuration before, during, or after flight. The convertible aircraft system includes a fuselage, a proximal flight assembly, a distal flight assembly, a support spar, and a tail assembly. The fuselage is the main structural body of the present invention. The proximal flight assembly and the distal flight assembly are the flight system of the present invention. The support spar provides an axis of rotation and a pole support for the proximal flight assembly and the distal flight assembly. The tail assembly provides stability during flight of the present invention. In more detail, the tail assembly may comprise at least one vertical stabilizer, at least one horizontal stabilizer, and at least one rudder in order to provide stability during flight of the present invention.

Abstract: The present application provides a model training, image processing method, device, storage medium, and program product relating to deep learning technology, which are able to screen auxiliary image data with image data for learning a target task, and further fuse the target image data and the auxiliary image data, so as to train a built and to-be-trained model with the fusion-processed fused image data. This implementation can increase the amount of data for training the model, and the data for training the model is determined is based on the target image data, which is suitable for learning the target task. Therefore, the solution provided by the present application can train an accurate target model even if the amount of target image data is not sufficient.

Abstract: A method, apparatus, device and readable medium for transfer learning in machine learning are provided. The method includes: constructing a target model according to the number of classes to be achieved by a target task and a duly-trained source model; obtaining a value of a regularized loss function of the corresponding target model and a value of a cross-entropy loss function of the target model, based on sets of training data in a training dataset of the target task; according to the value of the regularized loss function and the value of the cross-entropy loss function corresponding to each set of training data, updating parameters in the target model by a gradient descent method to implement the training of the target model. The above technical solution avoids excessive constraints on parameters in the prior art, thereby refraining from damaging the training effect of the source model on the target task.

Abstract: The present disclosure discloses a method and a device for locating an answer based on question and answer, in which the method includes: receiving a query sentence; parsing the query sentence, to generate a semantic parse tree corresponding to the query sentence; matching the semantic parse tree with a pre-established matching base, to obtain an alignment probability between each of candidate answers in the pre-established matching base and the semantic parse tree; and determining a final answer according to the alignment probability.

Abstract: The present disclosure provides a searching method and apparatus, a device and a non-volatile computer memory medium. According to embodiments of the present disclosure, it is possible to output the clustered search result under the potential demand by obtaining a search result according to the obtained query keyword, and then clustering the search result under the potential demand of the query keyword. Since the user might have demands in one or more aspects, clustering the search result corresponding to the query keyword under one or more potential demands of the query keyword can enable the user to easily obtain content in a class under a certain potential demand, and can effectively satisfy the user's relevant demands appearing during the search.

Abstract: Embodiments of the present invention provide a method, system and storage medium for implementing intelligent question answering. The method includes: receiving a query question; performing a semantic analysis of the question; performing corresponding search processing for the question based on a result of the semantic analysis, wherein the search processing includes search processing performed for the question by at least one of a semantic relationship mining system, a text library search system, a knowledge base search system, and a question and answer library search system; and returning an answer based on a result of the search processing. In this way, the accuracy of answers to the questions is effectively improved.

Abstract: The present disclosure discloses a method and a device for locating an answer based on question and answer, in which the method includes: receiving a query sentence; parsing the query sentence, to generate a semantic parse tree corresponding to the query sentence; matching the semantic parse tree with a pre-established matching base, to obtain an alignment probability between each of candidate answers in the pre-established matching base and the semantic parse tree; and determining a final answer according to the alignment probability.

Abstract: Embodiments of the present invention provide a method, system and storage medium for implementing intelligent question answering. The method includes: receiving a query question; performing a semantic analysis of the question; performing corresponding search processing for the question based on a result of the semantic analysis, wherein the search processing includes search processing performed for the question by at least one of a semantic relationship mining system, a text library search system, a knowledge base search system, and a question and answer library search system; and returning an answer based on a result of the search processing. In this way, the accuracy of answers to the questions is effectively improved.

Abstract: A magnetically actuated visual display board comprises a grid array of open ended chambers. Display configurations in each chamber utilize a magnetically responsive sliding element to move a display loop to alternate between two different display surfaces to create a desired visual array. A magnetic rod passing a chamber moves the sliding element within the chamber.

Abstract: A center foam strip has a series of transverse fold lines separating the strip into panels which fold onto each other in an accordion pattern for storage. A soft fabric waterproof cover material is attached to the center foam strip to hang down the sides of the vehicle to be closed in the vehicle doors shading the side windows. The cover material and center foam strip are sewn together to conform to the shape of the top of a particular vehicle with the foam strip covering the roof, windshield, and rear window. The center foam strip and surrounding soft fabric are held in place by the trunk lid, doors, and windshield wipers.

Abstract: A magnetically actuated visual display board comprises a grid array of open ended chambers. Display configurations in each chamber utilize a magnetically responsive sliding element to move a display loop to alternate between two different display surfaces to create a desired visual array. A magnetic rod passing a chamber moves the sliding element within the chamber.

Abstract: A self roll-up car cover has a built-in spring between waterproof sheets or a synthetic sheet with a built-in memory to return to a spiral shape when rolled out of the spiral configuration and then released. The rear half of the cover has an end rod and a central opening that allows it to be hooked over a trunk lock mechanism and kept secure when the trunk is then shut. The front half of the cover has a lead rope or belt with two hooks for securing under the hood of the vehicle or any other edged surface. The tension in the cover keeps it stretched out and the overlapping portion of the front and back covers maintains at least a portion of the cover elevated above the vehicle.