Abstract: Heterostructure and double-heterostructure trench-gate devices, in which the substrate and/or the body are constructed of a narrower-bandgap semiconductor material than the uppermost portion of the drift region. Fabrication most preferably uses a process where gate dielectric anneal is performed after all other high-temperature steps have already been done.

Abstract: Apparatuses, systems, and techniques to train one or more neural networks using stratified sampled training data parameters. In at least one embodiment, one or more stochastic training data parameters may be stratified sampled from one or more sampling ranges to compute a gradient for updating the one or more neural networks.

Abstract: In one example in accordance with the present disclosure, an electronic device is described. An example electronic device includes a processor and memory storing executable instructions that when executed cause the processor to receive an image of a polymer melt pool captured during crystallization of the polymer melt pool. The instructions also cause the processor to measure brightness of the polymer melt pool in the captured image. The instructions further cause the processor to determine crystallization information of the polymer melt pool based on the measured brightness of the polymer melt pool.

Abstract: A push-bending forming device for titanium alloy tubes is provided, which includes a die main body, a plurality of filling balls, a differential temperature assembly and a punch. The filling balls are configured to be filled in a tube blank, and are made of a metal material, with a heat-resistance temperature over 500° C. A diameter of the filling ball is less than an inner diameter of the tube blank. The punch includes a first pushing part and a second pushing part. A push-bending forming method using such device is also provided.

Abstract: An affine prediction method, a computing device and a non-transitory storage medium are disclosed. The method includes obtaining a current coding block comprising a plurality of sub-blocks arranged in an array; dividing the current coding block into at least two areas, wherein each area of the at least two areas comprises two or more of the sub-blocks; setting a corresponding control point motion vector (CPMV) for one or more control points in the each area, respectively; and affine-predicting motion vectors (MVs) of the two or more of the sub-blocks in the each area, using the corresponding CPMV.

Abstract: A coding algorithm encodes consecutive frames of a video sequence and realizes distributed Gradual Decoding Refresh. The consecutive frames include a first frame, a second frame, and a third frame, and each frame is composed of (X+Y+Z) columns of coding tree units. The algorithm includes: coding X columns of the first frame in an intra coding manner and coding the other columns of the first frame in an inter coding manner and/or the intra coding manner; coding X columns of the second frame in the inter coding manner, coding Y columns of the second frame in the intra coding manner, and coding Z columns of the second frame in the inter and/or intra coding manner(s); and coding X and Y columns of the third frame in the inter coding manner, and coding Z columns of the third frame in the intra coding manner. The X/Y/Z columns (intra-coded columns) are inconsecutive.

Abstract: A compact high-speed electrical connector and electronic system thereof. The connecter comprises first, second, and third mating interfaces configured to receive one or more cards, and a mounting interface configured to mount to a substrate. The first, second, and third mating interfaces extend in columns parallel to each other and perpendicular to a mating direction. The first, second, and third mating interfaces offset from each other in a mating direction. The mounting interface extends in the mating direction. Such a configuration enables direction interconnections between three components that support hybrid functions such as high-speed signals, power, and/or side bands, while satisfying a limited space provided by an electronic system.

Abstract: Examples of methods are described herein. In some examples, a method includes generating, using a compensation machine learning model after training, a compensated model based on a three-dimensional (3D) object model. In some examples, the compensation machine learning model is trained by generating candidate compensation plans and evaluating, using a deformation machine learning model, the candidate compensation plans. In some examples, the method includes adjusting the 3D object model based on the compensated model to produce an adjusted model.

Abstract: Compositions and methods for editing an endogenous NAC genes in plants are provided, for the improvement of traits of agronomic or commercial importance. Modifications of expression and activity of NAC genes are described, including editing endogenous NAC gene functional motifs and knocking out NAC gene function.

Abstract: The present provides an affine prediction method and related devices. The method may include: dividing a current coding block into multiple sub-blocks, and determining an initial prediction value of all pixels in each of the sub-blocks; dividing each of the sub-blocks into multiple fraction blocks, wherein at least one of the fraction blocks comprises at least two integer pixels; determining a motion vector difference and a gradient of each of the fraction blocks; calculating a pixel compensation value of each of the fraction blocks based on the motion vector difference value and the gradient; taking the pixel compensation value of each of the fraction blocks as a pixel compensation value of all pixels in each of the fraction blocks; and calculating a final prediction value of each pixel in the current coding block by the initial prediction value of each pixel and the pixel compensation value.

Abstract: A method for acquiring motion vectors includes: obtaining a number of related motion vectors of a current block, wherein the number of related motion vectors are motion vectors of a related block of the current block, a prediction mode of the related block is the same as a current prediction mode of the current block, an encoding has been completed; determining weights of at least part of the number of related motion vectors; and calculating a weighted motion vector of the current block based on at least part of the number of related motion vectors and the weights of at least part of the plurality of related motion vectors.

Abstract: Systems and method for determining a topic cohesion measurement between a content item and a hyperlinked landing page are presented. In one embodiment, a plurality of content item signals is generated for the content item and a corresponding plurality of signals are generated for the hyperlinked landing page. An analysis of the corresponding signals is conducted to determine a measurement of topic cohesion, a topic cohesion score, between the content item and the hyperlinked landing page. A cohesion predictor model is trained to generate the predictive topic cohesion score between an input content item and a hyperlinked landing page. Upon a determination that the topic cohesion score is less than a predetermined threshold, remedial actions are taken regarding the hyperlink of the content item. Alternatively, positive actions may be carried out, including promoting the content item to others, associating advertisements with the content item, and the like.

Abstract: A sub-block motion prediction method includes: obtaining a reference block, a co-located frame, and a reference frame of a current block; obtaining a motion vector of the current block, wherein the motion vector corresponds to a set of reference parameters; obtaining a co-located block of a sub-block based on the set of reference parameters, wherein a plurality of sub-blocks are obtained by dividing the current block; obtaining a candidate motion vector of each sub-block based on a motion vector of the co-located block; and obtaining a sub-block motion prediction result of the current block based on all candidate motion vectors of all sub-blocks and following a principle of consuming a minimal prediction cost.

Abstract: A touch control structure includes a plurality of touch signal lines in a peripheral area. A respective touch signal line includes a double-layer structure in a double-layer region and a single-layer structure in a single-layer region. The peripheral area includes a first sub-area on a first side, a second sub-area on a second side, a third sub-area on a third side, a fourth sub-area on a fourth side, of the touch control area. The first sub-area includes a side region, and one or more corner regions. The double-layer region and the single-layer region are in the first sub-area, the first sub-area has a first shortest width along a direction from the touch control area to the first sub-area, the first shortest width is greater than a shortest width of at least one of sub-areas of the peripheral area other than the first sub-area.

Abstract: Examples of methods are described herein. In some examples, a method includes determining a graph representation of a three-dimensional (3D) object. In some examples, the graph representation includes nodes and edges associated with the nodes. In some examples, each node includes a temperature profile attribute. In some examples, the method includes predicting, using a machine learning model, a deformation of the 3D object based on the graph representation.

Abstract: The present disclosure relates to a motion vector prediction method, a video coding method, and related apparatuses and devices. The motion vector prediction method includes: acquiring a candidate list of a current coding block in a preset prediction mode, where the candidate list includes a plurality of candidate motion vectors; performing a motion search of the plurality of candidate motion vectors in the candidate list to obtain an initial optimal motion vector of the current coding block in the preset prediction mode; performing motion compensation on the initial optimal motion vector by using each of a first preset number of pixel accuracies, to determine a final optimal motion vector of the current coding block, where the preset prediction mode comprises any one of a plurality of prediction modes, and pixel accuracies used in any one of the prediction modes are identical.