Abstract: A system for training an artificial intelligence (AI) model for an AI chip to implement an AI task may include an AI training unit to train weights of an AI model in floating point, a convolution quantization unit for quantizing the trained weights to a number of quantization levels, and an activation quantization unit for updating the weights of the AI model so that output of the AI model based at least on the updated weights are within a range of activation layers of the AI chip. The updated weights may be stored in fixed point and uploadable to the AI chip. The various units may be configured to account for the hardware constraints in the AI chip to minimize performance degradation when the trained weights are uploaded to the AI chip and expedite training convergence. Forward propagation and backward propagation may be combined in training the AI model.

Abstract: Disclosed are a method and system for correcting the precision of a magnetic levitation train traction system position control ring. The precision correction method comprises: step A, a speed measuring system collecting position-related information of a train; step B, sending the position-related information to a traction system through a signal system; and step C, the traction system carrying out closed-loop control on the position of the train according to the position-related information. The method further comprises step A1 before step A: checking the time for the speed measuring system, the signal system and the traction system, and adding timestamp information. According to the correcting method and system, the time is checked for a speed measuring system, a signal system and a traction system, and timestamp information is added, such that the influences of a delay and periodic random shaking are overcome, and the requirement of traction control of a medium-high speed magnetic levitation train is met.

Abstract: A rendered image is generated from a semiconductor device design file. The rendered image is segmented based on a grey level of the rendered image. Care areas are determined based on the segmenting. Defect inspection is performed in the care areas. This process can be performed on a wafer inspection tool that uses photon optics or electron beam optics.

Abstract: A glass composition with high Young's modulus and specific modulus, wherein the glass composition includes the following components by weight percentage: 30-46% of SiO2, 0-6% of B2O3, 10-30% of Al2O3, 4-20% of CaO, 2-15% of MgO, and 13-32% of Y2O3. The glass composition uses common chemical raw materials. By reasonably arranging the contents of each component, the glass is good in heat resistance, chemical stability and press-molding devitrification resistance performance, low in high-temperature viscosity, convenient in bubble elimination, with specific modulus above 34 and Young's modulus above 100 GPa, and is suitable for hard disk substrate manufacturing, semiconductor sealing and other fields.

Abstract: A sample characterization system is disclosed. In embodiments, the sample characterization system includes a controller communicatively coupled to an inspection sub-system, the controller including one or more processors configured to execute a set of program instructions stored in memory, the set of program instructions configured to cause the one or more processors to: acquire one or more target image frames of a sample; generate a target tensor with the one or more acquired target image frames; perform a first set of one or more decomposition processes on the target tensor to generate one or more reference tensors including one or more reference image frames; identify one or more differences between the one or more target image frames and the one or more reference image frames; and determine one or more characteristics of the sample based on the one or more identified differences.

Abstract: The present disclosure provides a modified montmorillonite self-repairing agent and a preparation method and use thereof, and belongs to the technical field of cement repairing materials. Montmorillonite is mixed with water, such that water is fully adsorbed between montmorillonite layers. The structure of montmorillonite is modified by supercritical CO2 treatment at a temperature of 50-60° C. and a pressure of 8-12 MPa, and the self-repairing efficiency of cement is improved by adding the modified montmorillonite into cement. Supercritical CO2 is adsorbed by montmorillonite, such that the montmorillonite is activated and an interlayer distance is increased to improve the repairing efficiency. When a crack is formed in cement, the adsorbed supercritical CO2 in montmorillonite is released into the crack and combined with positive ions to generate carbonate deposition, such that the crack is sealed and the self-repairing activity of the cement is improved.

Abstract: A rice cooker assembly uses machine learning models to identify and classify content in grain mixtures thereby to provide better automation of the cooking process. As one example, a rice cooker has a chamber storing grains. A camera is positioned to view an interior of the chamber. The camera captures images of the contents of the chamber. From the images, the machine learning model determines whether the contents of the chamber includes one type or multiple types of grain or whether the contents of the chamber includes any inedible objects. The machine learning model further classifies the one or more types of grains and inedible objects if any. The cooking process may be controlled accordingly. The machine learning model may be resident in the rice cooker or it may be accessed via a network.

Abstract: With the disclosed systems and methods for DRAM and 3D NAND inspection, an image of the wafer is received based on the output for an inspection tool. Geometric measurements of a design of a plurality of memory devices on the wafer are received. A care area with higher inspection sensitivity is determined based on the geometric measurements.

Abstract: Surface asperities, such as roughness characteristics, are reduced or otherwise mitigated via the control of surface regions including the asperities in different regimes. In accordance with various embodiments, the height of both high-frequency and low-frequency surface asperities is reduced by controlling characteristics of a surface region under a first regime to flow material from the surface asperities. A second regime is implemented to reduce a height of high-frequency surface asperities in the surface region by controlling characteristics of the surface region under a second regime to flow material that is predominantly from the high-frequency surface asperities, the controlled characteristics in the second regime being different than the controlled characteristics in the first regime. Such aspects may include, for example, controlling melt pools in each regime via energy pulses, to respectively mitigate/reduce the asperities.

Abstract: Provided herein are manufacturing methods of a metal matrix nanocomposite, comprising: providing a molten metal including a first reactant; providing a molten salt, including a second set of reactants and a diluting salt, over a surface of the molten metal; and maintaining the molten salt and the molten metal at a temperature sufficient to react the first reactant and the second set of reactants, such that nanostructures with controlled small sizes are formed adjacent to an interface between the molten salt and the molten metal, and are incorporated into the molten metal for mass manufacturing of metal matrix nanocomposite.

Abstract: Disclosed are an optical gradation system and method. The optical gradation system comprises: a first device configured to emit broad-spectrum linearly polarized light in a first polarization direction; a second device provided with at least one first region for splitting lights of different wavelengths mixed in the broad-spectrum linearly polarized light incident by the first device into emergent lights in different polarization directions without changing a beam propagation path; a third device configured to filter out linearly polarized light in a second polarization direction from the emergent lights from the second device in different polarization directions; and, a rotator configured to drive at least one of the first device, the second device and the third device to rotate, wherein the first device, the second device and the third device are arranged coaxially.

Abstract: In some embodiments, a system includes an artificial intelligence (AI) chip and a processor coupled to the AI chip and configured to receive an input image, crop the input image into a plurality of cropped images, and execute the AI chip to produce a plurality of feature maps based on at least a subset of the plurality of cropped images. The system may further merge at least a subset of the plurality of feature maps to form a merged feature map, and produce an output image based on the merged feature map. The cropping and merging operations may be performed according to a same pattern. The system may also include a training network configured to train weights of the CNN model in the AI chip in a gradient descent network. Cropping and merging may be performed over the training sample images in the training work in a similar manner.

Abstract: Systems and methods for detecting defects on a reticle are provided. One system includes computer subsystem(s) configured for performing at least one repeater defect detection step in front-end processing during an inspection process performed on a wafer having features printed in a lithography process using a reticle. The at least one repeater defect detection step performed in the front-end processing includes identifying any defects detected at corresponding locations in two or more test images by double detection and any defects detected by stacked defect detection as first repeater defect candidates. One or more additional repeater defect detections may be performed on the first repeater defect candidates to generate final repeater defect candidates and identify defects on the reticle from the final repeater defect candidates.

Abstract: Methods and systems for determining one or more alignment parameters for use in a process performed on a specimen are provided. One method includes determining measures of similarity between images generated by an imaging system for corresponding locations in each of two or more pairs of dies on a specimen and performing cluster analysis based on the determined measures of similarity to identify the images that are most similar to each other and to assign different subsets of the images that are most similar to each other to different die clusters, respectively. The method also includes separately determining one or more alignment parameters for two or more of the different die clusters. The one or more alignment parameters are used for aligning images generated by the imaging system for the specimen or another specimen to a common reference.

Abstract: Disclosed is a stone extraction basket, comprising a mesh basket (1), an outer tube (2) and an operation portion (10), wherein the mesh basket (1) is connected to the operation portion (10), and the operation portion (10) can control the mesh basket (1) to move; a proximal end of the outer tube (2) is connected to the operation portion (10), the outer tube axially extends from the proximal end to a distal end, and the distal end of the outer tube (2) is provided with a double lumen end cap (5); and a guide wire (4) passing through one lumen channel (B7) of the double lumen end cap (5), and the mesh basket (1) passes through the other lumen channel (A6) of the double lumen end cap (5). The stone extraction basket solves the existing technical problem of the mesh basket being hampered from removing a calculus and being withdrawn from the biliary tract because the guide wire is driven by the end cap at the distal end of the mesh basket.

Abstract: With the disclosed systems and methods for DRAM and 3D NAND inspection, an image of the wafer is received based on the output for an inspection tool. Geometric measurements of a design of a plurality of memory devices on the wafer are received. A care area with higher inspection sensitivity is determined based on the geometric measurements.

Abstract: Systems and methods increase the signal to noise ratio of optical inspection of wafers to obtain higher inspection sensitivity. The computed reference image can minimize a norm of the difference of the test image and the computed reference image. A difference image between the test image and a computed reference image is determined. The computed reference image includes a linear combination of a second set of images.

Abstract: Methods and systems for setting up inspection of a specimen are provided. One system includes one or more computer subsystems configured for acquiring a reference image for a specimen and modifying the reference image to fit the reference image to a design grid thereby generating a golden grid image. The one or more computer subsystems are also configured for storing the golden grid image for use in inspection of the specimen. The inspection includes aligning a test image of the specimen generated from output of an inspection subsystem to the golden grid image.

Abstract: Global and local alignment energies are used in an image contrast metric. The image contrast metric can be used to find optical targets. Some pixels from a gradient magnitude image and a context range image from an optical image can be used to determine the image contrast metric. A heatmap from the image contrast metrics across part of a wafer can then be used to make a list of targets. Upper and lower confidence values can be applied to rank the available targets.