Method and apparatus with neural network data input and output control

    公开(公告)号:US11790232B2

    公开(公告)日:2023-10-17

    申请号:US18095674

    申请日:2023-01-11

    Inventor: Hyung-Dal Kwon

    CPC classification number: G06N3/08 G06N3/04 H03M7/30

    Abstract: A neural network deep learning data control apparatus includes: a memory; an encoding circuit configured to receive a data sequence, generate a compressed data sequence in which consecutive invalid bits in a bit string of the data sequence are compressed into a single bit of the compressed data sequence, generate a validity determination sequence indicating a valid bit and an invalid bit in a bit string of the compressed data sequence, and write the compressed data sequence and the validity determination sequence to the memory; and a decoding circuit configured to read the compressed data sequence and the validity determination sequence from the memory, and determine a bit in the bit string of the compressed data sequence set for transmission to a neural network circuit, based on the validity determination sequence, such that the neural network circuit omits an operation with respect to non-consecutive invalid bits.

    Touch screen controller for increasing data processing speed and touch system including the same

    公开(公告)号:US10355737B2

    公开(公告)日:2019-07-16

    申请号:US16180871

    申请日:2018-11-05

    Abstract: A touch screen controller (TSC) includes: a front end circuit configured to send a control signal to a touch panel and to receive a touch signal from the touch panel; an algorithm processing circuit configured to process source data generated based on the touch signal according to a predetermined algorithm; a memory configured to store the source data and result data obtained as a result of processing the source data at the algorithm processing circuit; and a bus configured to transfer data among the front end circuit, the algorithm processing circuit, and the memory. The algorithm processing circuit includes: a buffer configured to temporarily store the source data or the result data and shared by at least two circuits; and a special function register (SFR) configured to store a setting value necessary for an operation of the algorithm processing circuit.

    Method and apparatus with neural network data input and output control

    公开(公告)号:US11580393B2

    公开(公告)日:2023-02-14

    申请号:US16900007

    申请日:2020-06-12

    Inventor: Hyung-Dal Kwon

    Abstract: A neural network deep learning data control apparatus includes: a memory; an encoding circuit configured to receive a data sequence, generate a compressed data sequence in which consecutive invalid bits in a bit string of the data sequence are compressed into a single bit of the compressed data sequence, generate a validity determination sequence indicating a valid bit and an invalid bit in a bit string of the compressed data sequence, and write the compressed data sequence and the validity determination sequence to the memory; and a decoding circuit configured to read the compressed data sequence and the validity determination sequence from the memory, and determine a bit in the bit string of the compressed data sequence set for transmission to a neural network circuit, based on the validity determination sequence, such that the neural network circuit omits an operation with respect to non-consecutive invalid bits.

    Touch screen controller for increasing data processing speed and touch system including the same

    公开(公告)号:US10141972B2

    公开(公告)日:2018-11-27

    申请号:US15393501

    申请日:2016-12-29

    Abstract: A touch screen controller (TSC) includes: a front end circuit configured to send a control signal to a touch panel and to receive a touch signal from the touch panel; an algorithm processing circuit configured to process source data generated based on the touch signal according to a predetermined algorithm; a memory configured to store the source data and result data obtained as a result of processing the source data at the algorithm processing circuit; and a bus configured to transfer data among the front end circuit, the algorithm processing circuit, and the memory. The algorithm processing circuit includes: a buffer configured to temporarily store the source data or the result data and shared by at least two circuits; and a special function register (SFR) configured to store a setting value necessary for an operation of the algorithm processing circuit.

    Method and apparatus with deep learning operations

    公开(公告)号:US12236336B2

    公开(公告)日:2025-02-25

    申请号:US17168457

    申请日:2021-02-05

    Inventor: Hyung-Dal Kwon

    Abstract: Disclosed is a method and apparatus with deep learning operations. A deep learning apparatus includes a processor, configured to support a plurality of different operation modes, including a systolic array having a plurality of multiplier accumulator (MAC) units, and a control circuit configured to respectively control, for each the plurality of different operation modes, select operations of the plurality of MAC units and data movements among the plurality of MAC units.

Patent Agency Ranking