QUANTUM INTERFERENCE BASED LOGIC DEVICES INCLUDING ELECTRON MONOCHROMATOR
    41.
    发明申请
    QUANTUM INTERFERENCE BASED LOGIC DEVICES INCLUDING ELECTRON MONOCHROMATOR 有权
    基于量子干扰的逻辑器件,包括电子单色器

    公开(公告)号:US20150123701A1

    公开(公告)日:2015-05-07

    申请号:US14478344

    申请日:2014-09-05

    Abstract: A logic device is provided which includes an electron monochromator. The electron monochromator includes a quantum dot disposed between first and second tunneling barriers, an emitter coupled to the first tunneling barrier, and a collector coupled to the second tunneling barrier. The logic device also includes a quantum interference device. The quantum interference device includes a source which is coupled to the collector of the electron monochromator.

    Abstract translation: 提供了一种包括电子单色仪的逻辑器件。 电子单色仪包括设置在第一和第二隧道势垒之间的量子点,耦合到第一隧道势垒的发射极和耦合到第二隧穿势垒的集电极。 逻辑器件还包括量子干涉器件。 量子干涉装置包括耦合到电子单色仪的集电极的源。

    CRYSTALLINE MULTIPLE-NANOSHEET III-V CHANNEL FETS
    42.
    发明申请
    CRYSTALLINE MULTIPLE-NANOSHEET III-V CHANNEL FETS 有权
    晶体多层纳米III-V通道FET

    公开(公告)号:US20150123215A1

    公开(公告)日:2015-05-07

    申请号:US14270690

    申请日:2014-05-06

    CPC classification number: H01L29/42392 H01L29/78681 H01L29/78696

    Abstract: A field effect transistor includes a body layer comprising a crystalline semiconductor channel region therein, and a gate stack on the channel region. The gate stack includes a crystalline semiconductor gate layer, and a crystalline semiconductor gate dielectric layer between the gate layer and the channel region. Related devices and fabrication methods are also discussed.

    Abstract translation: 场效应晶体管包括其中包括晶体半导体沟道区的主体层,以及沟道区上的栅叠层。 栅极堆叠包括晶体半导体栅极层和栅极层和沟道区之间的晶体半导体栅极介电层。 还讨论了相关设备和制造方法。

    Method of enabling sparse neural networks on memresistive accelerators

    公开(公告)号:US11816563B2

    公开(公告)日:2023-11-14

    申请号:US16409487

    申请日:2019-05-10

    CPC classification number: G06N3/08 G06F17/16

    Abstract: A method of storing a sparse weight matrix for a trained artificial neural network in a circuit including a series of clusters. The method includes partitioning the sparse weight matrix into at least one first sub-block and at least one second sub-block. The first sub-block includes only zero-value weights and the second sub-block includes non-zero value weights. The method also includes assigning the non-zero value weights in the at least one second sub-block to at least one cluster of the series of clusters of the circuit. The circuit is configured to perform matrix-vector-multiplication (MVM) between the non-zero value weights of the at least one second sub-block and an input vector during an inference process utilizing the artificial neural network. The sub-blocks containing all zero elements are power gated, thereby reducing overall energy consumption for inference.

    METHOD OF ENABLING SPARSE NEURAL NETWORKS ON MEMRESISTIVE ACCELERATORS

    公开(公告)号:US20200234114A1

    公开(公告)日:2020-07-23

    申请号:US16409487

    申请日:2019-05-10

    Abstract: A method of storing a sparse weight matrix for a trained artificial neural network in a circuit including a series of clusters. The method includes partitioning the sparse weight matrix into at least one first sub-block and at least one second sub-block. The first sub-block includes only zero-value weights and the second sub-block includes non-zero value weights. The method also includes assigning the non-zero value weights in the at least one second sub-block to at least one cluster of the series of clusters of the circuit. The circuit is configured to perform matrix-vector-multiplication (MVM) between the non-zero value weights of the at least one second sub-block and an input vector during an inference process utilizing the artificial neural network. The sub-blocks containing all zero elements are power gated, thereby reducing overall energy consumption for inference.

Patent Agency Ranking