Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.

Abstract: Provided herein is technology relating to molecular biological manipulation of genes and genomes and particularly, but not exclusively, to CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) methods, compositions, systems, and kits for improved genetic editing.

Abstract: A method includes: obtaining a push content request from an object; determining a push content set according to the push content request; detecting historical behavior feature data of the object for a push content in response to a behavior feature of the object for the push content satisfying a preset negative behavior condition; performing content-based matching for the push content set based on the historical behavior feature data to obtain a target push content set of the object; performing content sorting of the push content set and performing content-based matching of the push content of the object to obtain a target push content set of the object; and determining a target push content of the object from the target push content set, and transmitting the target push content to the object.

Abstract: Provided is a genetically modified non-human mammal that comprises an anchor DNA sequence inserted at an endogenous locus of a secretory milk protein gene, wherein the anchor DNA sequence comprises a site-specific recombinase recognition site. Also provided is a genetically modified non-human mammal that comprises a transgene inserted at an endogenous locus of a secretory milk protein gene, wherein the transgene encodes a secretory protein and is operably linked to the endogenous promoter of said secretory milk protein gene, and wherein the transgene is flanked by a pair of site-specific recombinase resulting sites. The genetically modified non-human mammals provided can be used for producing the secreted recombinant protein encoded by the transgene from the milk produced by the genetically modified non-human mammals.

Abstract: A positioning and rescue device for an unmanned underwater vehicle, comprising a battery (1), a switch (2), a protective resistor (3), an electromagnetic relay (4), a master control chip (5), a GPS positioning system (6), an igniter (7), a partition (8), an air bag (9), a shell (10), and a rope (11). The battery (1), the switch (2), the protective resistor (3), the electromagnetic relay (4) and the master control chip (5) are sequentially connected by means of wires to form a series circuit, and the igniter (7) and the GPS positioning system (6) are separately connected to the master control chip (5) by means of wires; two normally open contacts of a control loop of the electromagnetic relay (4) are respectively connected to two ends of a general power supply (15).

Abstract: This invention is in the field of medicinal pharmacology. In particular, the present invention relates to pharmaceutical agents which function as inhibitors of sodium-glucose cotransporter (SGLT) activity. The invention further relates to methods of treating and/or ameliorating symptoms related to cystic fibrosis (CF), comprising administering to a subject (e.g., a human patient) a composition comprising one or more pharmaceutical agents which function as inhibitors of SGLT activity.

Abstract: A method for machine learning-based account manager virtual assistant staging includes receiving a message and a classification, generating a staging record, generating a status using staging rules, generating an order when the message classification is order and the status is complete, and transmitting the order. An account manager virtual assistant staging system includes a processor and a memory storing instructions that cause the system to receive a message and a classification, generate a staging record, generate a status using staging rules, generate an order when the message classification is order and the status is complete, and transmit the order. A non-transitory computer readable medium contains program instructions that when executed, cause a computer to receive a message and a classification, generate a staging record, generate a status using staging rules, generate an order when the message classification is order and the status is complete, and transmit the order.

Abstract: A transformer simulation system and a measurement method for setting value simulation are disclosed. The transformer simulation system includes an oil tank, a heating device, a pump body, a first oil pipe, a second oil pipe and a first flow sensor. The heating device is disposed in the oil tank. The oil tank is provided with a first inlet and a first outlet, the first outlet is connected to the first oil pipe, and the first inlet is connected to the second oil pipe. An end of the first oil pipe away from the oil tank is a first mounting end for being connected to an oil inlet of a gas relay, and an end of the second oil pipe away from the oil tank is a second mounting end for being connected to an oil outlet of the gas relay. The pump body is connected to the first oil pipe or the second oil pipe, and the first flow sensor is disposed in the first oil pipe.

Abstract: A transformer simulation system and a measurement method for setting value simulation are disclosed. The transformer simulation system includes an oil tank, a heating device, a pump body, a first oil pipe, a second oil pipe and a first flow sensor. The heating device is disposed in the oil tank. The oil tank is provided with a first inlet and a first outlet, the first outlet is connected to the first oil pipe, and the first inlet is connected to the second oil pipe. An end of the first oil pipe away from the oil tank is a first mounting end for being connected to an oil inlet of a gas relay, and an end of the second oil pipe away from the oil tank is a second mounting end for being connected to an oil outlet of the gas relay. The pump body is connected to the first oil pipe or the second oil pipe, and the first flow sensor is disposed in the first oil pipe.

Abstract: A machine learning task automation method includes receiving a message, preprocessing the message to generate extracted data, analyzing the extracted data to generate classification data, analyzing the extracted data to generate information extraction; and generating and transmitting a response. A computing system includes a processor, and a memory including instructions that, when executed, cause the computing system to receive a message, preprocess the message to generate extracted data, analyze the extracted data to generate classification data, analyze the extracted data to generate information extraction data; and generate and transmit a response. A non-transitory computer readable medium containing program instructions that when executed, cause a computer to receive a message, preprocess the message to generate extracted data, analyze the extracted data to generate classification data, analyze the extracted data to generate information extraction data; and generate and transmit a response.

Abstract: An account manager virtual assistant computing system includes a processor and a memory storing instructions that when executed cause the system to receive a message, process the message using a quote classification machine learning model, generate a response including price information, and process the message using an information extraction machine deep learning model; and transmit the response. A non-transitory computer readable medium includes program instructions that when executed cause a computer to receive a message, process the message using a quote classification machine learning model, generate a response including price information, and process the message using an information extraction machine deep learning model; and transmit the response.

Abstract: A method for machine learning-based account manager virtual assistant message processing includes receiving a message of a user, preprocessing the message to generate a normalized data set, analyzing the normalized data using a classification machine learning model, analyzing the normalized data using a deep learning machine learning model, generating a response based on the output of the first and second machine learning models, and transmitting the response to the user. An account manager virtual assistant computing system includes a processor and a memory storing instructions that cause the account manager virtual assistant system to receive a message from a user, process the message using a first machine learning model, process the message using a second machine learning model, when the message is a request for quote, generate a response and transmit the response to the user.

Abstract: A method for machine learning-based account manager virtual assistant message processing includes receiving a message of a user, preprocessing the message to generate a normalized data set, analyzing the normalized data using a classification machine learning model, analyzing the normalized data using a deep learning machine learning model, generating a response based on the output of the first and second machine learning models, and transmitting the response to the user. An account manager virtual assistant computing system includes a processor and a memory storing instructions that cause the account manager virtual assistant system to receive a message from a user, process the message using a first machine learning model, process the message using a second machine learning model, when the message is a request for quote, generate a response and transmit the response to the user.

Abstract: Provided is a genetically modified non-human mammal that comprises an anchor DNA sequence inserted at an endogenous locus of a secretory milk protein gene, wherein the anchor DNA sequence comprises a site-specific recombinase recognition site. Also provided is a genetically modified non-human mammal that comprises a transgene inserted at an endogenous locus of a secretory milk protein gene, wherein the transgene encodes a secretory protein and is operably linked to the endogenous promoter of said secretory milk protein gene, and wherein the transgene is flanked by a pair of site-specific recombinase resulting sites. The genetically modified non-human mammals provided can be used for producing the secreted recombinant protein encoded by the transgene from the milk produced by the genetically modified non-human mammals.

Abstract: An optical lens for laser marking includes a first lens (L1), a second lens (L2), and a third lens (L3), which are successively coaxially arranged along a transmission direction of incident light, wherein the first lens (L1) and the second lens (L2) are meniscus lenses, and the third lens (L3) is a biconvex lens; wherein the first lens (L1) has a first surface (S1) and a second surface (S2), the second lens (L2) has a third surface (S3) and a fourth surface (S4), the third lens (L3) has a fifth surface (S5) and a sixth surface (S6); the first surface (S1) to the sixth surface (S6) are successively arranged along the transmission direction of the incident light; wherein radii of curvature of the first surface to the sixth surface are ?47±5% mm, ?, ?218±5% mm, ?81±5% mm, 778±5% mm, and ?142±5% mm, respectively; wherein central thicknesses of the first lens, the second lens, and the third lens are 4±5% mm, 15±5% mm, and 18±5% mm, respectively.

Abstract: An optical lens comprising a first lens (L1), a second lens (L2), a third lens (L3), a fourth lens (L4), and a fifth lens (L5) that are sequentially arranged on a common optical axis in the transmission direction of an incident light. Both the first lens (L1) and the fifth lens (L5) are negative meniscus lenses. Both the second lens (L2) and the third lens (L3) are positive meniscus lenses. The fourth lens (L4) is a biconvex lens. The optical lens is applicable in an optical system of a laser processing apparatus. When an employed processing wavelength is different from a monitoring wavelength, imaging color differences in a monitoring system can thus be eliminated, specifically, when a wavelength in the infrared range is employed as a laser processing wavelength while a red wavelength serves as the monitoring wavelength, improved imaging effects are provided in the monitoring system, thus ensuring the quality of laser processing.

Abstract: Provided herein are pharmaceutical compositions comprising valproic acid and methods of use thereof for the treatment and/or prevention of myocardial ischemia and/or infarction.

Abstract: A multi-purpose vehicle smart monitoring system and method. The monitoring system normally works in a standby manner for saving power in a smart sleep mode. When it is sensed that abnormalities occur inside, outside or around a vehicle, the monitoring system will start an automatic monitoring equipment, and notify a vehicle owner, so that the vehicle owner learns in a timely manner instant information about the vehicle and a peripheral condition of the vehicle anytime and anywhere. The vehicle owner also can operate the multi-purpose vehicle smart monitoring system so as to learn a real-time condition of and around the vehicle, and further can realize a plurality of functions, such as safe driving assistance and emergency assistance during driving.

Abstract: A telecentric optical lens (100), comprising a first lens (L1) to a fifth lens (L5) arranged sequentially and co-axially along the direction of transmission of the incident light ray; the first lens (L1) is a plano-concave negative lens; the second lens (L2) is a meniscus negative lens; the third lens (L3) is a meniscus negative lens; the fourth lens (L4) is a biconvex positive lens; and the fifth lens (L5) is a plano-concave negative lens; the arrangement and design parameters of the first lens (L1) to the fifth lens (L5) of the telecentric optical lens (100) allow said telecentric optical lens (100) to achieve a telecentric effect, and simultaneously satisfy achromatic and relatively large aperture requirements.

Abstract: An optical lens for laser marking includes a first lens (L1), a second lens (L2), and a third lens (L3), which are successively coaxially arranged along a transmission direction of incident light, wherein the first lens (L1) and the second lens (L2) are meniscus lenses, and the third lens (L3) is a biconvex lens; wherein the first lens (L1) has a first surface (S1) and a second surface (S2), the second lens (L2) has a third surface (S3) and a fourth surface (S4), the third lens (L3) has a fifth surface (S5) and a sixth surface (S6); the first surface (S1) to the sixth surface (S6) are successively arranged along the transmission direction of the incident light; wherein radii of curvature of the first surface to the sixth surface are ?47±5% mm, ?, ?218±5% mm, ?81±5% mm, 778±5% mm, and ?142±5% mm, respectively; wherein central thicknesses of the first lens, the second lens, and the third lens are 4±5% mm, 15±5% mm, and 18±5% mm, respectively.