Generating a training model based on feedback

    公开(公告)号:US11132618B2

    公开(公告)日:2021-09-28

    申请号:US15372821

    申请日:2016-12-08

    Abstract: A method and apparatus for generating a training model based on feedback are provided. The method for generating a training model based on feedback, includes calculating an eigenvector of a sample among a plurality of samples; obtaining scores granted by a user for one or more of the plurality of samples in a round, obtaining scores granted by the user for a first number of samples; obtaining scores granted by the user for a second number of samples in response to detecting, based on the eigenvector, an inconsistency between the scores granted by the user for the first number of samples; and generating a training model based on the scores granted by the user for the first and second numbers of samples. A corresponding apparatus is also provided.

    Facilitation of automatic adjustment of a braking system

    公开(公告)号:US11046293B2

    公开(公告)日:2021-06-29

    申请号:US16294122

    申请日:2019-03-06

    Abstract: Systems and methods for facilitating an automatic adjustment of a braking system is provided. In one example, a computer-implemented method can comprise generating, by a system operatively coupled to a processor, a braking curve model based on braking usage pattern data corresponding to one or more vehicles. The computer-implemented method can also comprise adjusting, by the system, a supplemental braking component of the first vehicle based on a simulation of one or more braking components corresponding to the one or more vehicles, wherein the one or more braking components is represented by the braking curve model.

    Ultra low parasitic inductance integrated cascode GaN devices

    公开(公告)号:US10991722B2

    公开(公告)日:2021-04-27

    申请号:US16355008

    申请日:2019-03-15

    Abstract: One silicon MOSFET transistor, which is used as the VThreshold control, and a GaN power HEMT are integrated on a single die to enable a fully integrated depletion-mode power device. GaN area is created on a silicon substrate and GaN FETs are built in the GaN area. Outside of the GaN area, silicon transistors such as switch MOSFETs are built. Front end of line or back end of line metal connections are then made to create interconnections among the GaN FET and the silicon transistor. The short physical proximity of the silicon transistor and GaN HEMT significantly reduces the parasitic resistance and inductance between them. Thus, high speed signals are able to travel from the silicon transistor to the GaN HEMT with a higher frequency and lower distortion, without creating overshoot voltage when there is large parasitic inductance. Therefore, the cascode device can operate at a higher switching frequency.

    Image sensor with analog sample and hold circuit control for analog neural networks

    公开(公告)号:US10708522B2

    公开(公告)日:2020-07-07

    申请号:US16100249

    申请日:2018-08-10

    Abstract: According to one or more embodiments of the present invention, an image processing system includes a cross-point synapse array that includes multiple row wires, multiple column wires, and multiple cross-point devices, a cross-point device at each intersection of the row wires and the column wires. The image processing system further includes an image sensor array that includes multiple pixel unit circuits, each pixel unit circuit is connected to a corresponding row wire of the cross-point synapse array, wherein the pixel unit circuit generates a voltage output based on an input light. The image processing system further includes a pixel unit controller that adjusts an exposure time of the pixel unit circuits based on voltage outputs from the pixel unit circuits respectively.

    Tape backup in large scale distributed systems

    公开(公告)号:US10579295B2

    公开(公告)日:2020-03-03

    申请号:US15293888

    申请日:2016-10-14

    Abstract: A computer-implemented method includes identifying a source data object of a distributed computing environment. The distributed computing environment includes two or more storage nodes. The source data object exists as two or more slices. At least one of the slices is replicated on at least two storage nodes. The computer-implemented method further includes associating the source data object with a tape. The tape is written by a tape drive controlled from the distributed computing environment. The computer-implemented method further includes copying the source data object to the tape by, for each source slice of the two or more slices, in sequence: selecting a source node of the two or more storage nodes whereon the source slice is replicated, mounting the tape drive to the source node, appending the source slice to the tape, and unmounting the tape drive. A corresponding computer program product and computer system are also disclosed.

Patent Agency Ranking