Abstract: In various embodiments, a data optimization application mitigates scoring inaccuracies in subjective quality experiments. In operation, the data optimization application generates a model that includes a first set of individual scores and a first set of parameters. The first set of parameters includes a first subjective score set and a first set of subjective factor sets. The data optimization application performs one or more optimization operations on the first set of parameters to generate a second set of parameters. The second set of parameters includes a second subjective score set and a second set of subjective factor sets, wherein the second subjective score set compensates for at least a first subjective factor set included in the second set of subjective factor sets.

Abstract: A refrigerator production line and a method for upgrading a refrigerator software of an assembled refrigerator on the refrigerator production line are provided. The refrigerator production line includes a conveying device for conveying an assembled refrigerator; a scanning device for scanning and confirming an identification code of the assembled refrigerator and an address code of a communication module in the assembled refrigerator; and a local server for saving the identification code of the assembled refrigerator and the address code of the communication module and further acquiring, in response to a firmware upgrade request, a firmware upgrading package for the assembled refrigerator that corresponds to the identification code. After the assembled refrigerator is powered on, the communication module receives the firmware upgrade package, and a control module upgrades a public version software program according to the firmware upgrade package.

Abstract: Service providers can be identified to fulfill service requests of a network-based service. A network system is configured to generate, based on historical data associated with the network-based service, a machine-learned service provider optimization (MLSPO) model for generating service provider optimizations. The optimizations can include action recommendations that optimize one or more service metrics. The MLSPO model can be a reinforcement learning model generated by performing a plurality of simulations utilizing one or more virtual agents. A provider device of a service provider can transmit a set of data to the network system that indicates a current location of the service provider. Based on the current location and the MLSPO model, the network system can generate service provider optimizations. Optimization data can be transmitted to the provider device so that the provider device can display information corresponding to the service provider optimizations.

Abstract: Systems, computer-readable media, and methods are disclosed for parallel data processing for service function chains with network functions spanning multiple servers. An example system includes a first server hosting a first network function of a service function chain, a second server hosting a second network function of the service function chain, a mirror function deployed in a first switch to replicate a plurality of packets received by the system and to send respective copies of the plurality of packets to the first network function and to at least one of the second network function and a third network function of the service function chain, and a merge function deployed in a second switch to merge respective outputs of the first network function and the at least one of the second network function and the third network function.

Abstract: The invention relates to an aerosol separator for separating aerosols in a blow-by gas, which comprises at least one feed opening for feeding a blow-by gas provided with aerosols, an outlet opening for discharging a blow-by gas cleaned of aerosols, at least one outflow opening for releasing aerosols removed from the blow-by gas, and which is arranged between a flow path of the at least one feed opening and the outlet opening, and a plurality of projections for configuring the flow path running from the at least one feed opening to the outflow opening in a meandering manner, in order to decelerate the blow-by gas which enters the feed opening and which is provided with aerosols, in terms of its flow rate. The separator is characterized in that the projections in their end region, around which blow-by gas flows, have a portion with increasing projection width.

Abstract: User activity pertaining to a product is automatically analyzed. Based, at least in part, on the analyzing the user activity pertaining to the product, a temporal logging level model can be derived. The temporal logging level model can determine a first logging level to be applied to the product. The temporal logging level model also can determine a duration of time for which the first logging level is to be applied to the product. Application of the first logging level to the product for the determined duration of time can be initiated.

Abstract: A balanced photodetector may include: a balanced photodiode including a first photodiode and a second photodiode coupled with one another at a common node, wherein the first photodiode has a first effective responsivity and the second photodiode has as second effective responsivity; and a control circuit configured to set an operating parameter of the balanced photodiode to compensate for a difference between the first effective responsivity and the second effective responsivity.

Abstract: This invention relates to a cosmetic composition comprising: (A) a film forming composition that can produce a tack free surface comprising: (I) a hydrosilylation reaction product of: (a) a silicone resin having the formula: (R1R22SiO1/2)w(R1R2SiO2/2)x(R2SiO3/2)y(SiO4/2)z wherein R2 is a C1 to C10 hydrocarbon group free of aliphatic unsaturation, R1 is R2 or an alkenyl group, w is from 0.3 to 0.6, x is from 0.0 to 0.2, y is 0, z is from 0.4 to 0.7, w+x+y+z=1.0, and w+z is from 0.85 to 1.

Abstract: Aspects of the invention include a computer-implemented method that includes converting runtime data points that are indicative of an influence of the runtime data points on a robustness and performance of a computing system. The runtime data points are clustered, wherein each cluster respectively represents a temporal state of the computing system. Each cluster is translated into a three-dimensional representation based on a probability density of the runtime data points of each cluster. A time-based vector is generated, where the vector describes a transition from a first three-dimensional representation to a second three-dimensional representation. Each three-dimensional representation traversed by the time-based vector represents a respective state of the computing system. The time-based vector is compared with a baseline vector. An anomaly alert is issued based at least in part on the comparison.

Abstract: A pump assembly, fluid machinery, and a heat exchange device. The pump assembly includes: a lower flange; a lower friction-reducing ring; a cylinder, the lower friction-reducing ring being disposed inside the cylinder, and the lower flange being disposed below the cylinder; and a piston assembly arranged inside the cylinder and includes a piston sleeve and a piston slidably arranged inside the piston sleeve. The lower friction-reducing ring has a central hole. A position-limiting protrusion is disposed on a surface of the piston sleeve, which faces the lower flange. The limiting protrusion extends into the central hole of the lower friction-reducing ring, fits and is limited by the lower flange.

Abstract: A method for correcting an error in a text, an electronic device, and a storage medium are provided. The method includes: obtaining an original text; obtaining a training text by preprocessing the original text; extracting a plurality of feature vectors corresponding to each word in the training text; obtaining an input vector by processing the plurality of feature vectors; obtaining a target text by inputting the input vector into a text error correction model; and adjusting parameters of the text error correction model based on a difference between the target text and the original text.

Abstract: A method and apparatus for training models in machine translation, an electronic device and a storage medium are disclosed, which relates to the field of natural language processing technologies and the field of deep learning technologies. An implementation includes mining similar target sentences of a group of samples based on a parallel corpus using a machine translation model and a semantic similarity model, and creating a first training sample set; training the machine translation model with the first training sample set; mining a negative sample of each sample in the group of samples based on the parallel corpus using the machine translation model and the semantic similarity model, and creating a second training sample set; and training the semantic similarity model with the second sample training set.

Abstract: In various embodiments, a tunable VMAF application reduces an amount of influence that image enhancement operations have on perceptual video quality estimates. In operation, the tunable VMAF application computes a first value for a first visual quality metric based on reconstructed video content and a first enhancement gain limit. The tunable VMAF application computes a second value for a second visual quality metric based on the reconstructed video content and a second enhancement gain limit. Subsequently, the tunable VMAF application generates a feature value vector based on the first value for the first visual quality metric and the second value for the second visual quality metric. The tunable VMAF application executes a VMAF model based on the feature value vector to generate a tuned VMAF score that accounts, at least in part, for at least one image enhancement operation used to generate the reconstructed video content.

Abstract: A method includes receiving, by a server, a data transmission of video content destined for a data storage device. The method includes generating, by the server, a content signature by application of a wavelet transform to a plurality of frames of a time sequence of frames of the video content. The method includes selecting, by the server based on content metadata for the video content, particular prohibited content signatures from a plurality of prohibited content signatures. Each prohibited content signature of the particular prohibited content signatures corresponds to a wavelet transform of a plurality of frames of prohibited content. The method also includes, in response to a first determination that a first prohibited content signature of the particular prohibited content signatures matches the content signature, preventing transmission of the video content from the server to the data storage device.

Abstract: The present disclosure provides an end-to-end model training method and apparatus, which relates to a field of artificial intelligence technologies. The method includes: obtaining training data containing a plurality of training samples, in which the plurality of training samples include an original sequence, a target sequence and a corresponding tag list, the tag list includes importance tags in the target sequence and avoidance tags corresponding to the importance tags, and the avoidance tags are irrelevant tags corresponding to the importance tags; and adopting the training data to train a preset end-to-end model until a value of a preset optimization target function is smaller than a preset threshold.

Abstract: Some embodiments of a disclosure provide a piston limiting structure, a compressor, and a heat exchange apparatus. The piston limiting structure includes: a cylinder, having a piston hole perpendicular to an axial direction of the cylinder and penetrating the cylinder, wherein a projection of the piston hole in a penetrating direction is circular; a piston, disposed in the piston hole and slid in the piston hole, wherein a side wall of the piston is provided with a limiting surface, and the limiting surface does not penetrate two ends of the side wall of the piston along an axial length of the piston; and a limiting member, wherein the cylinder is provided with a limiting hole penetrating from an outer wall of the cylinder to the piston hole, the limiting member is mounted in the limiting hole , and the limiting member abuts against the limiting surface.

Abstract: Provided is a piston limiting structure, including: a cylinder, a piston, and a flange provided with a limiting piece, the cylinder has a piston hole perpendicular to an axial direction of the cylinder and penetrating through the cylinder, and a projection of the piston hole in a penetrating direction is circular; the piston is disposed in the piston hole in a form-fit manner and is slid in the piston hole in a reciprocating manner, a side wall of the piston is provided with a thrust groove, a bottom surface of the thrust groove forms a thrust surface on the side wall of the piston, and the thrust groove does not penetrate through two ends of the side wall of the piston along an axial length of the piston; and the limiting piece abuts against the thrust surface.

Abstract: Systems and methods for predicting and/or detecting cardiac events based on real-time biomedical signals are discussed herein. In various embodiments, a machine learning algorithm may be utilized to predict and/or detect one or more medical conditions based on obtained biomedical signals. For example, the systems and methods described herein may utilize ECG signals to predict and detect cardiac events. In various embodiments, patterns identified within a signal may be assigned letters (i.e., encoded as distributions of letters). Based on the known morphology of a signal, states within the signal may be identified based on the distribution of letters in the signal. When applied in the in-vehicle environment, drivers or passengers within the vehicle may be alerted when an individual within the vehicle is, or is about to, experience a cardiac event.

Abstract: A method for text translation includes obtaining a text to be translated; and inputting the text to be translated into a text translation model. The trained text translation model divides the text to be translated into a plurality of semantic units, determines N semantic units before a current semantic unit among the plurality of semantic units as local context semantic units, determines M semantic units before the local context semantic units as global context semantic units, and generates a translation result of the current semantic unit based on the local context semantic units and the global context semantic units. N is an integer, and M is an integer.