Abstract: A method and system for text-to-video generation. The method includes receiving a text input, generating a representation frame based on the text input using a model trained on text-image pairs, generating a set of frames based on the representation frame and a first frame rate, interpolating the set of frames to a higher frame rate, generating a first video based on the interpolated set of frames, increasing a resolution of the first video based on a first and second super-resolution model, and generating an output video based on a result of the super-resolution models.

Abstract: A port departure service processing method, apparatus and system. The method includes: determining the current master control system for executing a port departure service; during the process of executing the port departure service, if an error is detected in operating data of the current master control system, switching the unfinished port departure service to another normally operating system so as to continue execution; if the execution of the port departure service is not completed and after it is detected that the operating data of the current master control system returns to normal, switching to the current master control system to continue to execute the remaining port departure service; and after switching from the current master control system to the other system or switching from the other system back to the current master control system, correcting inconsistent data before and after switching.

Abstract: The disclosure discloses a composition containing melatonin as an effective ingredient, and its application in improving DHI production performance and improving milk quality. The composition containing melatonin as an effective ingredient according to the present disclosure can significantly reduce the somatic cell count in milk and improve milk quality. Under the conditions of different seasons, ages, and lactation periods, the administration of the composition containing melatonin as an active ingredient according to the present disclosure can significantly reduce the somatic cell score of cow milk (p<0.01), increase the protein content and lactose content in milk, and meanwhile reduce the fat content and urea nitrogen content in milk. In summary, the composition containing melatonin as an effective ingredient according to the present disclosure can improve the quality of milk and increase the efficiency of dairy cow breeding.

Abstract: A first node obtains local processing time separately corresponding to a plurality of second nodes and/or uploading time of local processing results separately corresponding to the plurality of second nodes. The first node soils sequence numbers of the plurality of second nodes based on the local processing time and/or the uploading time, to obtain a first sequence. The first node determines a second sequence based on the first sequence, where the second sequence includes some or all second node sequence numbers in the first sequence. The first node schedules second nodes corresponding to the sequence numbers included in the second sequence.

Abstract: Systems and methods for predicting a risk of cardiovascular disease (CVD) from one or more fundus images are disclosed. Fundus images associated with an individual are processed to determine whether fundus images are of sufficient quality. The fundus images of sufficient quality are processed to identify fundus images belonging to a single eye. A plurality of risk contributing factor sets of CNNs (RCF CNN) are configured to output an indicator of probability of the presence of a different risk contributing factor in each of the one or more fundus images. At least one of the RCF CNNs is configured in a jury system model having a plurality of jury member CNNs, each being configured to output a probability of a different feature in the one or more fundus images. The outputs of the jury member CNNs are processed to determine the indicator of probability of the presence of the risk contributing factor output by the RCF CNN.

Abstract: Provided is a method for reducing methane production in animal stomachs by adding melatonin to gastric juice of the animal or feeding the animal with melatonin. In the method of the present disclosure, 10?7 mol/L to 10?3 mol/L of melatonin can reduce the methane production from the in vitro gastric fermentation fluid, feeding melatonin of 8.0 to 35.0 mg/kg/day can reduce the methane production from the animal respiration from day 7 of feeding melatonin. The method of the present disclosure can reduce greenhouse gas emissions of animals, control environmental pollution and realize low-carbon farming.

Abstract: A visual guidance-based mobile game system (001) comprises a terminal (100) and a gamepad (200) connected with the terminal (100). The terminal (100) comprises a user interface module configured to receive an eye movement image and output a mobile game response. The mobile game system (001) further comprises a visual intelligent guidance module (210) and a data processing module (220) that are connected with the user interface module (120). The eye tracker converts the eye movement image into the eye tracking information. The visual intelligent guidance module (210) is configured to transmit the eye tracking information to the data processing module (220). The data processing module (220) is configured to determine a responsive instruction corresponding to the eye tracking information according to the eye tracking information, so as to guide a control point on a screen of the terminal according to the responsive instruction and complete the mobile game response.

Abstract: A training method for a robust neural network based on feature matching is provided in this disclosure, which includes following steps. Step A, a first stage model is initialized. The first stage model includes a backbone network, a feature matching module and a fullple loss function. Step B, the first stage model is trained by using original training data to obtain a second stage model. Step C, the second stage model is attacked so as to generate PGD adversarial samples of the original training data, and the second stage model is trained again with the generated adversarial samples and the original training data. Step D, training parameters are adjusted and the second stage model is trained again, and parameters for which the model has highest accuracy on an original test set are saved.

Abstract: A soundproof door system includes a chute; a door body, where a bottom of the door body is inserted into the chute and able to slide along the chute; a first roller disposed on the bottom of the door body and supported on a bottom wall of the chute; a first soundproof plate provided with a first end connected to the door body and a second end abutting against an inner wall of the chute; a second soundproof plate provided with a third end connected to the door body and a fourth end abutting against the inner wall of the chute; and the first end and the third end are respectively located on two sides of the door body, and the second end and the fourth end are respectively located on both sides of the door body.

Abstract: A training method for a robust neural network based on feature matching is provided in this disclosure, which includes following steps. Step A, a first stage model is initialized. The first stage model includes a backbone network, a feature matching module and a fullple loss function. Step B, the first stage model is trained by using original training data to obtain a second stage model. Step C, the second stage model is attacked so as to generate PGD adversarial samples of the original training data, and the second stage model is trained again with the generated adversarial samples and the original training data. Step D, training parameters are adjusted and the second stage model is trained again, and parameters for which the model has highest accuracy on an original test set are saved.

Abstract: A contribution identification method for noise at a boundary of an urban substation in the present disclosure includes the following steps: describing site elements and surrounding environmental elements of an urban substation; measuring and recording a background noise value at each boundary of the urban substation; determining an orientation of a specific sound source of the urban substation, and setting a corresponding measurement point at each boundary of the urban substation; obtaining a spectrum of each specific sound source; analyzing a contribution of each intra-substation sound source of the urban substation to noise at the boundary of the substation; analyzing a noise level of each sound source measurement point in the urban substation; recording measured data during noise measurement, and correcting a noise measurement result; and generating a test report.

Abstract: Disclosed is an optimal power flow acquiring method for regional distribution network of small hydropower groups based on deep learning, which specifically includes the following steps: generating required data sets by adopting continuous power flow and power flow equation calculation methods; the data set is randomly divided into training data (80 percent) and test data (20 percent); training the built convolutional neural network model with training data to learn the mapping relationship between load and generator output power; inputting test data, and directly obtaining PG and QG from the trained convolutional neural network; and solving residual variables Vi and ?i with traditional power flow solver. The application can accelerate the solving speed of the optimal power flow problem with higher prediction accuracy.

Abstract: Methods and systems for image alignment are provided. One method includes selecting three or more salient feature points for use in test image to reference image alignment by applying a selected salient feature point detection method to at least a reference image generated for the specimen. The method also includes detecting the three or more salient feature points in the test image and the reference image and aligning the detected three or more salient feature points in the test image to the detected three or more salient feature points in the reference image. The method further includes aligning remaining portions of the test image to remaining portions of the reference image based on results of the previous aligning step.

Abstract: Disclosed is a composition for increasing pregnancy rate in ruminants exposed to ovsynch and timed artificial insemination, and a preparation method and application thereof. The effective component of the composition is melatonin with a concentration of 10-20 mg/mL. The present disclosure proves through animal experiments that the use of the composition of the present disclosure in combination with the last injection of GnRH in an ovsynch and timed artificial insemination protocol for a ruminant can regulate the secretion of reproductive hormones in an animal body, increase the levels of luteinizing hormone (LH) and progesterone in pregnant females, and improve ovum quality and facilitate ovulation, thus increasing the pregnancy rate in ruminants exposed to ovsynch and timed artificial insemination.

Abstract: The disclosure discloses a composition containing melatonin as an effective ingredient, and its application in improving DHI production performance and improving milk quality. The composition containing melatonin as an effective ingredient according to the present disclosure can significantly reduce the somatic cell count in milk and improve milk quality. Under the conditions of different seasons, ages, and lactation periods, the administration of the composition containing melatonin as an active ingredient according to the present disclosure can significantly reduce the somatic cell score of cow milk (p<0.01), increase the protein content and lactose content in milk, and meanwhile reduce the fat content and urea nitrogen content in milk. In summary, the composition containing melatonin as an effective ingredient according to the present disclosure can improve the quality of milk and increase the efficiency of dairy cow breeding.

Abstract: The Machine Learning Portfolio Simulating and Optimizing Apparatuses, Methods and Systems (“MLPO”) transforms machine learning simulation request, decision tree ensembles training request, expected returns calculation request, portfolio construction request, predefined scenario construction request, portfolio returns visualization request inputs via MLPO components into machine learning simulation response, decision tree ensembles training response, expected returns calculation response, portfolio construction response, predefined scenario construction response, portfolio returns visualization response outputs. A portfolio return computation request configured to specify simulated market scenarios generated using neural networks and a set of filters is obtained. Constituent portfolio securities of a portfolio are determined. The simulated market scenarios are filtered based on the set of filters. Expected returns for the constituent portfolio securities are retrieved.

Abstract: The Machine Learning Portfolio Simulating and Optimizing Apparatuses, Methods and Systems (“MLPO”) transforms machine learning simulation request, decision tree ensembles training request, expected returns calculation request, portfolio construction request, predefined scenario construction request, portfolio returns visualization request inputs via MLPO components into machine learning simulation response, decision tree ensembles training response, expected returns calculation response, portfolio construction response, predefined scenario construction response, portfolio returns visualization response outputs. A portfolio return computation request configured to specify simulated market scenarios generated using multi-variate mixture datastructures and a set of filters is obtained. Constituent portfolio securities of a portfolio are determined. The simulated market scenarios are filtered based on the set of filters. Expected returns for the constituent portfolio securities are retrieved.

Abstract: The Machine Learning Portfolio Simulating and Optimizing Apparatuses, Methods and Systems (“MLPO”) transforms machine learning simulation request, decision tree ensembles training request, expected returns calculation request, portfolio construction request, predefined scenario construction request, portfolio returns visualization request inputs via MLPO components into machine learning simulation response, decision tree ensembles training response, expected returns calculation response, portfolio construction response, predefined scenario construction response, portfolio returns visualization response outputs. User selection of simulated market scenarios generated using neural networks is obtained. A range of unfiltered simulated market factor values for each market factor is determined. Customized market factors are updated based on a user modification. A range of allowable values for each customized market factor is determined.

Abstract: The Machine Learning Portfolio Simulating and Optimizing Apparatuses, Methods and Systems (“MLPO”) transforms machine learning simulation request, decision tree ensembles training request, expected returns calculation request, portfolio construction request, predefined scenario construction request, portfolio returns visualization request inputs via MLPO components into machine learning simulation response, decision tree ensembles training response, expected returns calculation response, portfolio construction response, predefined scenario construction response, portfolio returns visualization response outputs. An asset return metrics calculation request datastructure is obtained. The number of sessions to utilize for calculating asset return metrics data is determined.

Abstract: The Machine Learning Portfolio Simulating and Optimizing Apparatuses, Methods and Systems (“MLPO”) transforms machine learning simulation request, decision tree ensembles training request, expected returns calculation request, portfolio construction request, predefined scenario construction request, portfolio returns visualization request inputs via MLPO components into machine learning simulation response, decision tree ensembles training response, expected returns calculation response, portfolio construction response, predefined scenario construction response, portfolio returns visualization response outputs. User selection of simulated market scenarios generated using multi-variate mixture datastructures is obtained. A range of unfiltered simulated market factor values for each market factor is determined. Customized market factors are updated based on a user modification. A range of allowable values for each customized market factor is determined.