TWI800226B - 用於執行神經網路計算的類比電路的校準方法及裝置 - Google Patents

用於執行神經網路計算的類比電路的校準方法及裝置 Download PDF

Info

Publication number
TWI800226B
TWI800226B TW111102245A TW111102245A TWI800226B TW I800226 B TWI800226 B TW I800226B TW 111102245 A TW111102245 A TW 111102245A TW 111102245 A TW111102245 A TW 111102245A TW I800226 B TWI800226 B TW I800226B
Authority
TW
Taiwan
Prior art keywords
calibration
neural network
network computing
analog circuits
analog
Prior art date
Application number
TW111102245A
Other languages
English (en)
Other versions
TW202230225A (zh
Inventor
陳柏亨
李佳達
張櫂閔
鄭志崇
黃瀚韜
叢培貴
魏軍浩
陳明裕
Original Assignee
新加坡商聯發科技（新加坡）私人有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 新加坡商聯發科技（新加坡）私人有限公司 filed Critical 新加坡商聯發科技（新加坡）私人有限公司
Publication of TW202230225A publication Critical patent/TW202230225A/zh
Application granted granted Critical
Publication of TWI800226B publication Critical patent/TWI800226B/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
    • G06F7/38—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation
    • G06F7/48—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices
    • G06F7/544—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices for evaluating functions by calculation
    • G06F7/5443—Sum of products
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F7/00—Methods or arrangements for processing data by operating upon the order or content of the data handled
    • G06F7/38—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation
    • G06F7/48—Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state device; using unspecified devices
    • G06F7/483—Computations with numbers represented by a non-linear combination of denominational numbers, e.g. rational numbers, logarithmic number system or floating-point numbers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G06N3/065—Analogue means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Neurology (AREA)
  • Nonlinear Science (AREA)
  • Image Analysis (AREA)
  • Analogue/Digital Conversion (AREA)
TW111102245A 2021-01-20 2022-01-19 用於執行神經網路計算的類比電路的校準方法及裝置 TWI800226B (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US202163139463P 2021-01-20 2021-01-20
US63/139,463 2021-01-20
US17/569,771 US20220230064A1 (en) 2021-01-20 2022-01-06 Calibration of analog circuits for neural network computing
US17/569,771 2022-01-06

Publications (2)

Publication Number Publication Date
TW202230225A TW202230225A (zh) 2022-08-01
TWI800226B true TWI800226B (zh) 2023-04-21

Family

ID=82405228

Family Applications (1)

Application Number Title Priority Date Filing Date
TW111102245A TWI800226B (zh) 2021-01-20 2022-01-19 用於執行神經網路計算的類比電路的校準方法及裝置

Country Status (3)

Country Link
US (1) US20220230064A1 (zh)
CN (1) CN114819051A (zh)
TW (1) TWI800226B (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024068298A1 (en) * 2022-09-26 2024-04-04 Interdigital Ce Patent Holdings, Sas Mixing analog and digital neural networks implementations in video coding processes

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106133763A (zh) * 2014-03-27 2016-11-16 高通股份有限公司 可塑性突触管理
CN111279366A (zh) * 2017-10-24 2020-06-12 国际商业机器公司 人工神经网络的训练
TW202029065A (zh) * 2018-09-07 2020-08-01 南韓商三星電子股份有限公司 積體電路、神經網路處理器與裝置以及資料處理方法
US20200311535A1 (en) * 2019-03-25 2020-10-01 Northeastern University Self-powered analog computing architecture with energy monitoring to enable machine-learning vision at the edge
TW202042233A (zh) * 2019-05-02 2020-11-16 美商超捷公司 用於深度學習人工神經網路中之類比神經記憶體的輸出陣列神經元轉換與校正
CN112101539A (zh) * 2020-11-18 2020-12-18 南京优存科技有限公司 一种存算一体电路及人工智能芯片

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106133763A (zh) * 2014-03-27 2016-11-16 高通股份有限公司 可塑性突触管理
CN111279366A (zh) * 2017-10-24 2020-06-12 国际商业机器公司 人工神经网络的训练
TW202029065A (zh) * 2018-09-07 2020-08-01 南韓商三星電子股份有限公司 積體電路、神經網路處理器與裝置以及資料處理方法
US20200311535A1 (en) * 2019-03-25 2020-10-01 Northeastern University Self-powered analog computing architecture with energy monitoring to enable machine-learning vision at the edge
TW202042233A (zh) * 2019-05-02 2020-11-16 美商超捷公司 用於深度學習人工神經網路中之類比神經記憶體的輸出陣列神經元轉換與校正
CN112101539A (zh) * 2020-11-18 2020-12-18 南京优存科技有限公司 一种存算一体电路及人工智能芯片

Also Published As

Publication number Publication date
CN114819051A (zh) 2022-07-29
TW202230225A (zh) 2022-08-01
US20220230064A1 (en) 2022-07-21

Similar Documents

Publication Publication Date Title
EP4163831A4 (en) METHOD AND DEVICE FOR DISTILLING NEURAL NETWORK
EP4145251A4 (en) METHOD FOR INTERACTION WITH AN ELECTRONIC DEVICE FOR SKIN INSPECTION, AND ELECTRONIC DEVICE
EP4135226A4 (en) METHOD AND APPARATUS FOR ADJUSTING NEURONAL NETWORK
TWI800226B (zh) 用於執行神經網路計算的類比電路的校準方法及裝置
EP3916678A4 (en) ELECTRONIC DEVICE AND METHOD FOR CONTROLLING A GRAPHIC OBJECT OF THE ELECTRONIC DEVICE
EP4180991A4 (en) METHOD AND DEVICE FOR THE DISTILLATION OF A NEURAL NETWORK
EP4131077A4 (en) METHOD AND DEVICE FOR NEURONAL NETWORK OPTIMIZATION
EP4101136A4 (en) INITIAL ACCESS METHOD AND ASSOCIATED DEVICE
EP4167127A4 (en) METHOD FOR IDENTIFYING THE HEALTH STATUS OF AN OBJECT, AND ASSOCIATED ELECTRONIC DEVICE
EP3953869A4 (en) AI MODEL TRAINING METHOD AND ELECTRONIC APPARATUS
EP4170548A4 (en) METHOD AND DEVICE FOR CONSTRUCTING A NEURONAL NETWORK
EP4273754A4 (en) NEURAL NETWORK TRAINING METHOD AND RELATED APPARATUS
EP4131982A4 (en) UPDATE RATE SETTING METHOD AND ELECTRONIC DEVICE
EP4192185A4 (en) COMPUTER SUPPORT APPLICATION METHOD AND APPARATUS
EP4262121A4 (en) TRAINING METHOD FOR NEURONAL NETWORK AND ASSOCIATED APPARATUS
EP4141757A4 (en) EXPOSURE DOSE MEASUREMENT SIMULATION APPARATUS AND METHOD
EP4113266A4 (en) ELECTRONIC DEVICE AND CONTROL METHOD
EP4206909A4 (en) APPLICATION PROCESSING METHOD AND ASSOCIATED APPARATUS
EP3975528A4 (en) Control method of electronic device and electronic device
EP4044125A4 (en) ELECTRONIC DEVICE AND ITS CONTROL METHOD
EP4250685A4 (en) NETWORK CONFIGURATION METHOD AND APPARATUS FOR AN INTERNET OF THINGS DEVICE
EP4206624A4 (en) VIBRATION DISTRIBUTION MEASURING DEVICE AND CORRESPONDING METHOD
EP4163827A4 (en) METHOD AND DEVICE FOR ACQUIRING AN ARTIFICIAL NEURAL NETWORK
EP4102338A4 (en) ELECTRONIC DEVICE AND METHOD FOR PROVIDING USER LOCATION
EP4163418A4 (en) COATING DEVICE AND CORRESPONDING COATING METHOD