Abstract: An embodiment of the present invention is directed toward machine learning to produce results encompassing a new output. A machine learning model is trained to determine a candidate output from among a plurality of candidate outputs. First embeddings associated with the plurality of candidate outputs are generated from a first set of training data by an intermediate layer of the trained machine learning model. Second embeddings associated with a new candidate output are generated from a second set of training data by the intermediate layer of the trained machine learning model. A third embedding is determined for input data by the intermediate layer of the trained machine learning model. A resulting candidate output for the input data is predicted from a group of the plurality of candidate outputs and the new candidate output based on distances for the third embedding to the first and second embeddings.

Abstract: A system and method for management of heterogeneous bare metal servers is disclosed. The system includes target bare metal servers coupled to a network and an administrative server. The administrative server executes a framework for managing the bare metal servers. The framework includes unified APIs for performance of management functions. A recognition engine receives identification information from the server and selects a driver from a driver library based on the identification information. Each of the drivers in the driver library is associated with a different type of bare metal server. A driver loader loads the selected driver to translate the unified APIs to native APIs on the bare metal server to perform the management function on the server.

Abstract: Techniques are described that enable automatic emergency braking (AEB) for a path-crossing target when a collision between a host vehicle and the target that is deemed imminent. Based on whether an acceleration of the host vehicle is above a threshold. Based on the acceleration, and, optionally, a location of the target relative to a crossing path (e.g., whether a portion of the target is within a suppression zone), an AEB system of the host vehicle is either activated or not activated, for example, suppressed. This suppression of the AEB system may include gating or nulling an AEB activation signal to prevent an emergency braking event. By managing the AEB system in a path-crossing scenario, many common false-positive AEB events (warnings, alerts, and/or braking) may be avoided. Furthermore, intentional vehicle maneuvers that comply with normal driving etiquette or rules can still be allowed for operator and passenger comfort, without risking safety.

Abstract: A connector plug, abutting structure of the same, and electrical connection mechanism having the same are provided. The connector plug includes a connection body and an outer casing member. The outer casing member is disposed outside the connection body and includes an inner casing, an outer casing movably fitted to the inner casing from outside, and a resilient component acting between the inner casing and the outer casing. At least one hook hole is disposed at the front end of the outer casing. At least one engaging arm is disposed on the inner casing and extends forward to form a hook portion. At least one engaging hole is disposed on a receptacle. When a plug is plugged into the receptacle, the hook portion becomes engaged with the engaging hole to achieve connection.

Abstract: A connector plug, abutting structure of the same, and electrical connection mechanism having the same are provided. The connector plug includes a connection body and an outer casing member. The outer casing member is disposed outside the connection body and includes an inner casing and an outer casing enclosing the inner casing. At least one engaging arm is disposed on the inner casing and has a pressing portion protruding from the outer casing and an engaging portion extending forward from the pressing portion and hanging on the sidewall of an insertion port of the connection body. At least one engaging hole is defined on a receptacle. When a plug is plugged into the receptacle, the engaging portion becomes engaged with the engaging hole to achieve connection.

Abstract: A package structure includes a semiconductor die, an insulating encapsulant, a first redistribution layer, a second redistribution layer, antenna elements and a first insulating film. The insulating encapsulant is encapsulating the at least one semiconductor die, the insulating encapsulant has a first surface and a second surface opposite to the first surface. The first redistribution layer is disposed on the first surface of the insulating encapsulant. The second redistribution layer is disposed on the second surface of the insulating encapsulant. The antenna elements are located over the second redistribution layer. The first insulating film is disposed in between the second redistribution layer and the antenna elements, wherein the first insulating film comprises a resin rich region and a filler rich region, the resin rich region is located in between the filler rich region and the second redistribution layer and separating the filler rich region from the second redistribution layer.

Abstract: A demand forecasting method and a demand forecasting apparatus are provided. A preliminary prediction amount corresponding to a part number is obtained based on historical demand data. A demand probability of the part number is calculated based on the preliminary prediction amount. A prediction demand amount corresponding to the part number is obtained based on the historical demand data, the preliminary prediction amount and the demand probability.

Abstract: A method for receiving a full training data set including a plurality of individual training data set, dividing the plurality of individual training sets into N classes, where N is an integer greater than three, dividing the N classes into M full data classes and N-M partial data classes, performing training to obtain a trained fixed size machine learning (ML) classification model and a trained in-class confidence model, outputting a first set of prediction value(s) based on the performance of training, distributing each class of the N classes of individual training data sets to a different node of a distributed machine learning system; and outputting, from the nodes of the distributed machine learning system, a second set of prediction value(s) for each class of the N classes.

Abstract: A shift register and a gate driver circuit are provided. The shift register includes an input unit, an output unit, an electrostatic discharge unit and a reset unit. The input unit provides an input signal. The output unit is coupled to the input unit and a gate output terminal. The output unit outputs an output signal through the gate output terminal according to the input signal. The electrostatic discharge unit is coupled to the output unit. After the gate output terminal outputs the output signal, the electrostatic discharge unit pulls down a voltage of the gate output terminal according to a low gate voltage. The reset unit is coupled to the input unit and the output unit. After the electrostatic discharge unit pulls down the voltage of the gate output terminal, the reset unit resets a voltage of a bootstrap node.

Abstract: A window includes a window frame and a screen mounting assembly. The window frame is formed with two rail slots. The screen mounting assembly is adapted to be connected to a window screen and includes a roller unit, a sliding unit and an interconnecting unit. The roller unit includes a roller seat. The sliding unit is slidable along the rail slots and defines an engaging groove. The interconnecting unit is operable between a disengaging state where the interconnecting unit is connected to the roller seat, and an engaging state where the interconnecting unit is separated from the roller seat and inserted into the engaging groove so as to engage the sliding unit, such that movement of the sliding unit along the rail slots results in winding and unwinding of the window screen.

Abstract: An oil filter removal tool configured for removal of a filter cartridge is provided, including: a main body, including a driving portion configured to be connected with a driven portion of the filter cartridge so that the main body is rotatable with the filter cartridge; a cover portion, connected with the main body, including an opening corresponding to the driving portion, the opening being configured for assembling of the driven portion of the filter cartridge therewithin; wherein the main body has a first flexibility, the cover portion has a second flexibility, and the second flexibility is greater than the first flexibility.

Abstract: A method for producing a dental adaptive element for alleviating recession of gingiva around an alveolar socket of a patient includes: obtaining images of a tooth; creating a 3D model having a crown part and a root part based on the images; obtaining a boundary curve on the 3D model between the crown part and the root part; on the root part of the 3D virtual model, obtaining an imaginary surface that protrudes in a direction from the crown part toward the root part; extracting a sub-model from the 3D model by removing a portion of the root part that extends between the imaginary surface and an end of the root part; and producing a dental adaptive element according to the sub-model.

Abstract: A method of detecting or monitoring process electrical charge produced during fabrication of an integrated circuit (IC) on a semiconductor wafer includes fabricating a process charge detection circuit on or in the semiconductor wafer, including: a victim isolation well, a gate oxide disposed on or in the victim isolation well, an aggressor isolation well electrically connected with the victim isolation well via the gate oxide, a victim antenna electrically connected with the victim isolation well and together with the victim isolation well defining a victim RC time constant, and an aggressor antenna electrically connected with the aggressor isolation well and together with the aggressor isolation well defining an aggressor RC time constant that is different from the victim RC time constant. Process charge is detected using the process charge detection circuit. The detecting comprises measuring an electrical parameter of the gate oxide.

Abstract: The present disclosure relates to a semiconductor device including a substrate and a pair of spacers on the substrate. Each spacer of the pair of spacers includes an upper portion having a first width and a lower portion under the upper portion and having a second width different from the first width. The semiconductor device further includes a gate structure between the pair of spacers. The gate structure has an upper gate length and a lower gate length that is different from the upper gate length.

Abstract: A lamination chuck for lamination of film materials includes a support layer and a top layer. The top layer is disposed on the support layer. The top layer includes a polymeric material having a Shore A hardness lower than a Shore hardness of a material of the support layer. The top layer and the support layer have at least one vacuum channel formed therethrough, vertically extending from a top surface of the top layer to a bottom surface of the support layer.

Abstract: A fingerprint sensing module suitable for receiving a sensing light beam and an electronic device is provided. The fingerprint sensing module includes a sensing element, a light transmitting layer disposed on the sensing element, a micro-lens layer disposed on the light transmitting layer, and a first light shielding layer disposed in the light transmitting layer and including multiple first openings arranged in an array. Positions of the first openings in odd-numbered rows are the same, and those in even-numbered rows are the same. Positions the first openings in the odd-numbered rows are different from those in the even-numbered rows. A sensing light beam includes multiple first light beams incident to a part of a sensing unit in a first transmission direction and multiple second light beams incident to another part of the sensing unit in a second transmission direction. The first transmission direction is different from the second transmission direction.

Abstract: A semiconductor device and a manufacturing method thereof are provided. The semiconductor device includes a substrate, a first well, a second well and doped regions. The substrate has heavily doped and lightly doped regions over the heavily doped region. The first wells are disposed in the lightly doped region and arranged as an array. The first wells have a conductive type opposite to a conductive type of the heavily doped and lightly doped regions. The second well is disposed in the substrate over the lightly doped region, and has an active region defined by an isolation structure. The first wells are overlapped with the second well. Top ends of the first wells are lower than a bottom end of the second well. The doped regions are separately located in the active region, and have a conductive type opposite to a conductive type of the second well.

Abstract: A passive optical network device comprising a casing, printed circuit board, and fiber optic transceiver system is provided. The fiber optic transceiver system comprises a fiber optic components device, fiber optic transceiver, and RF connector. During operation, the fiber optic components device converts optical signals from the fiber optic transceiver to digital signals, and then transmits the converted digital signals to external electronic systems via the hot-pluggable transceiver connection interface. The fiber optic components device converts digital signals from the external electronic systems to optical signals, and then transmits the optical signals to other external electronic systems via the fiber optic transceiver. The RF connector transmits RF signals from additional external electronic systems to the external electronic systems via the hot-pluggable transceiver connection interface.

Abstract: An electronic device and an image signal processing method of removing background noise based on spatial frequency are provided. The electronic device includes an image sensor and a processor. The image sensor senses an image of an object. The processor executes image signal processing operations of: receiving a composite image generated by the image sensor; transforming the composite image from a spatial domain to a frequency domain to obtain a composite frequency domain component; eliminating a background component, corresponding to a frequency domain noise position, from a second frequency domain component to obtain a clear frequency domain component representative of the image signal; and performing subsequent processing according to the clear frequency domain component to generate the image signal for the electronic device.