DK3361422T3 - Implemenerting af neurale netværk i aritmetiske computersystemer med fast komma - Google Patents
Implemenerting af neurale netværk i aritmetiske computersystemer med fast komma Download PDFInfo
- Publication number
- DK3361422T3 DK3361422T3 DK18156667.0T DK18156667T DK3361422T3 DK 3361422 T3 DK3361422 T3 DK 3361422T3 DK 18156667 T DK18156667 T DK 18156667T DK 3361422 T3 DK3361422 T3 DK 3361422T3
- Authority
- DK
- Denmark
- Prior art keywords
- arthmetic
- committee
- implementation
- fixed
- computer systems
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/10—Interfaces, programming languages or software development kits, e.g. for simulating neural networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F5/00—Methods or arrangements for data conversion without changing the order or content of the data handled
- G06F5/01—Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
- G06F5/012—Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising in floating-point computations
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
- G06N3/0495—Quantised networks; Sparse networks; Compressed networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06N3/0499—Feedforward networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2207/00—Indexing scheme relating to methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F2207/38—Indexing scheme relating to groups G06F7/38 - G06F7/575
- G06F2207/3804—Details
- G06F2207/3808—Details concerning the type of numbers or the way they are handled
- G06F2207/3812—Devices capable of handling different types of numbers
- G06F2207/3824—Accepting both fixed-point and floating-point numbers
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2207/00—Indexing scheme relating to methods or arrangements for processing data by operating upon the order or content of the data handled
- G06F2207/38—Indexing scheme relating to groups G06F7/38 - G06F7/575
- G06F2207/48—Indexing scheme relating to groups G06F7/48 - G06F7/575
- G06F2207/4802—Special implementations
- G06F2207/4818—Threshold devices
- G06F2207/4824—Neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Neurology (AREA)
- Nonlinear Science (AREA)
- Image Analysis (AREA)
- Feedback Control In General (AREA)
- Advance Control (AREA)
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US15/432,842 US10650303B2 (en) | 2017-02-14 | 2017-02-14 | Implementing neural networks in fixed point arithmetic computing systems |
| US15/476,769 US9836692B1 (en) | 2017-02-14 | 2017-03-31 | Implementing neural networks in fixed point arithmetic computing systems |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| DK3361422T3 true DK3361422T3 (da) | 2022-01-31 |
Family
ID=60452224
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| DK18156667.0T DK3361422T3 (da) | 2017-02-14 | 2018-02-14 | Implemenerting af neurale netværk i aritmetiske computersystemer med fast komma |
Country Status (9)
| Country | Link |
|---|---|
| US (3) | US10650303B2 (da) |
| EP (1) | EP3361422B1 (da) |
| CN (2) | CN116415637A (da) |
| DE (1) | DE102018103303A1 (da) |
| DK (1) | DK3361422T3 (da) |
| GB (1) | GB2561669A (da) |
| HK (1) | HK1259158A1 (da) |
| TW (4) | TWI823571B (da) |
| WO (1) | WO2018152205A1 (da) |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10417525B2 (en) * | 2014-09-22 | 2019-09-17 | Samsung Electronics Co., Ltd. | Object recognition with reduced neural network weight precision |
| US10650303B2 (en) * | 2017-02-14 | 2020-05-12 | Google Llc | Implementing neural networks in fixed point arithmetic computing systems |
| US10467796B2 (en) | 2017-04-17 | 2019-11-05 | Intel Corporation | Graphics system with additional context |
| CN109961133B (zh) * | 2017-12-14 | 2020-04-24 | 中科寒武纪科技股份有限公司 | 集成电路芯片装置及相关产品 |
| EP3789871B1 (en) * | 2017-12-27 | 2023-06-07 | Cambricon Technologies Corporation Limited | Integrated circuit chip device |
| EP3640863B1 (en) * | 2018-02-13 | 2021-10-27 | Shanghai Cambricon Information Technology Co., Ltd | Computation device and method |
| KR102808577B1 (ko) | 2018-04-27 | 2025-05-15 | 삼성전자주식회사 | 뉴럴 네트워크의 파라미터들을 양자화하는 방법 및 장치 |
| KR102740541B1 (ko) | 2018-09-03 | 2024-12-09 | 삼성전자주식회사 | 인공 신경망 시스템 및 인공 신경망의 고정 소수점 제어 방법 |
| CN109284761B (zh) * | 2018-09-04 | 2020-11-27 | 苏州科达科技股份有限公司 | 一种图像特征提取方法、装置、设备及可读存储介质 |
| KR102816285B1 (ko) | 2018-09-07 | 2025-06-02 | 삼성전자주식회사 | 뉴럴 프로세싱 시스템 |
| US11663464B2 (en) * | 2018-09-20 | 2023-05-30 | Kneron (Taiwan) Co., Ltd. | Deep neural network with low-precision dynamic fixed-point in reconfigurable hardware design |
| KR102775183B1 (ko) * | 2018-11-23 | 2025-03-04 | 삼성전자주식회사 | 뉴럴 네트워크 연산 수행을 위한 뉴럴 네트워크 장치, 뉴럴 네트워크 장치의 동작 방법 및 뉴럴 네트워크 장치를 포함하는 애플리케이션 프로세서 |
| CN109670582B (zh) * | 2018-12-28 | 2021-05-07 | 四川那智科技有限公司 | 一种全定点化神经网络的设计方法 |
| CN113272831B (zh) * | 2019-01-18 | 2024-07-23 | 日立安斯泰莫株式会社 | 神经网络的缩简装置 |
| CN110008952B (zh) * | 2019-03-26 | 2021-06-15 | 深兰科技(上海)有限公司 | 一种目标识别方法及设备 |
| WO2021077283A1 (zh) * | 2019-10-22 | 2021-04-29 | 深圳鲲云信息科技有限公司 | 神经网络计算压缩方法、系统及存储介质 |
| US11601134B2 (en) | 2020-01-10 | 2023-03-07 | Robert Bosch Gmbh | Optimized quantization for reduced resolution neural networks |
| US20220156567A1 (en) * | 2020-11-13 | 2022-05-19 | Mediatek Inc. | Neural network processing unit for hybrid and mixed precision computing |
| CN112702600B (zh) * | 2020-12-29 | 2022-04-12 | 南京大学 | 一种图像编解码神经网络分层定点化方法 |
| CN113342346B (zh) * | 2021-05-18 | 2022-03-25 | 北京百度网讯科技有限公司 | 深度学习框架的算子注册方法、装置、设备和存储介质 |
| TWI778822B (zh) * | 2021-09-30 | 2022-09-21 | 鴻海精密工業股份有限公司 | 基於神經網路模型的影像處理方法、電子設備及儲存介質 |
| EP4626036A1 (en) * | 2024-03-27 | 2025-10-01 | GN Hearing A/S | Hearing device with low power neural network |
Family Cites Families (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW305998B (en) | 1996-06-14 | 1997-05-21 | Ind Tech Res Inst | Weighted sum type of artificial neural network with reconfigurable structure and bit-serial I/O mode |
| US8443169B2 (en) * | 2005-03-28 | 2013-05-14 | Gerald George Pechanek | Interconnection network connecting operation-configurable nodes according to one or more levels of adjacency in multiple dimensions of communication in a multi-processor and a neural processor |
| US7917333B2 (en) | 2008-08-20 | 2011-03-29 | Caterpillar Inc. | Virtual sensor network (VSN) based control system and method |
| US9015204B2 (en) | 2009-07-15 | 2015-04-21 | Hewlett-Packard Development Company, L.P. | Onboarding resources to an identity management system |
| KR101084581B1 (ko) * | 2009-11-02 | 2011-11-17 | 전자부품연구원 | 고정 소수점 지수함수 연산 방법, 장치 및 기록매체 |
| US8805914B2 (en) * | 2010-06-02 | 2014-08-12 | Maxeler Technologies Ltd. | Method and apparatus for performing numerical calculations |
| TWI525558B (zh) | 2011-01-17 | 2016-03-11 | Univ Nat Taipei Technology | Resilient high - speed hardware reverse transfer and feedback type neural network system |
| US9104473B2 (en) * | 2012-03-30 | 2015-08-11 | Altera Corporation | Conversion and compression of floating-point and integer data |
| CN102981854A (zh) | 2012-11-16 | 2013-03-20 | 天津市天祥世联网络科技有限公司 | 基于浮点数运算内联函数库的神经网络优化方法 |
| US20160019027A1 (en) * | 2014-07-15 | 2016-01-21 | Qualcomm Incorporated | Vector scaling instructions for use in an arithmetic logic unit |
| WO2016039651A1 (en) | 2014-09-09 | 2016-03-17 | Intel Corporation | Improved fixed point integer implementations for neural networks |
| US20170061279A1 (en) * | 2015-01-14 | 2017-03-02 | Intel Corporation | Updating an artificial neural network using flexible fixed point representation |
| US10373050B2 (en) | 2015-05-08 | 2019-08-06 | Qualcomm Incorporated | Fixed point neural network based on floating point neural network quantization |
| US11244225B2 (en) | 2015-07-10 | 2022-02-08 | Samsung Electronics Co., Ltd. | Neural network processor configurable using macro instructions |
| JP6745019B2 (ja) * | 2015-10-29 | 2020-08-26 | 株式会社Preferred Networks | 情報処理装置及び情報処理方法 |
| CN105760933A (zh) | 2016-02-18 | 2016-07-13 | 清华大学 | 卷积神经网络的逐层变精度定点化方法及装置 |
| US10650303B2 (en) * | 2017-02-14 | 2020-05-12 | Google Llc | Implementing neural networks in fixed point arithmetic computing systems |
-
2017
- 2017-02-14 US US15/432,842 patent/US10650303B2/en active Active
- 2017-03-31 US US15/476,769 patent/US9836692B1/en active Active
-
2018
- 2018-02-14 TW TW111135158A patent/TWI823571B/zh active
- 2018-02-14 DE DE102018103303.6A patent/DE102018103303A1/de not_active Withdrawn
- 2018-02-14 TW TW107105678A patent/TWI728230B/zh active
- 2018-02-14 EP EP18156667.0A patent/EP3361422B1/en active Active
- 2018-02-14 DK DK18156667.0T patent/DK3361422T3/da active
- 2018-02-14 CN CN202211709740.3A patent/CN116415637A/zh active Pending
- 2018-02-14 TW TW110115931A patent/TWI787803B/zh active
- 2018-02-14 CN CN201810152235.0A patent/CN108427991A/zh active Pending
- 2018-02-14 GB GB1802414.1A patent/GB2561669A/en not_active Withdrawn
- 2018-02-14 TW TW107202461U patent/TWM573476U/zh unknown
- 2018-02-14 WO PCT/US2018/018184 patent/WO2018152205A1/en not_active Ceased
-
2019
- 2019-01-23 HK HK19101205.8A patent/HK1259158A1/zh unknown
-
2020
- 2020-03-26 US US16/830,922 patent/US11868864B2/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| HK1259158A1 (zh) | 2019-11-29 |
| US20180232626A1 (en) | 2018-08-16 |
| GB2561669A (en) | 2018-10-24 |
| TWM573476U (zh) | 2019-01-21 |
| EP3361422B1 (en) | 2021-12-08 |
| WO2018152205A1 (en) | 2018-08-23 |
| EP3361422A1 (en) | 2018-08-15 |
| CN108427991A (zh) | 2018-08-21 |
| TWI728230B (zh) | 2021-05-21 |
| TWI823571B (zh) | 2023-11-21 |
| DE102018103303A1 (de) | 2018-08-16 |
| CN116415637A (zh) | 2023-07-11 |
| US9836692B1 (en) | 2017-12-05 |
| TWI787803B (zh) | 2022-12-21 |
| US11868864B2 (en) | 2024-01-09 |
| US20200226445A1 (en) | 2020-07-16 |
| TW201835818A (zh) | 2018-10-01 |
| TW202131233A (zh) | 2021-08-16 |
| US10650303B2 (en) | 2020-05-12 |
| TW202301201A (zh) | 2023-01-01 |
| GB201802414D0 (en) | 2018-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DK3361422T3 (da) | Implemenerting af neurale netværk i aritmetiske computersystemer med fast komma | |
| IL269230A (en) | Systems and methods for thermal regulation | |
| IL272113B (en) | Methods and systems for environmental credit scoring | |
| IL262787B (en) | Systems and methods incorporating a neural network and a forward physical model for semiconductor applications | |
| IL270450B1 (en) | Self-righting systems and related components and methods | |
| EP3679486C0 (en) | CLOUD-BASED IMAGE ACCESS SYSTEMS AND METHODS | |
| EP3245965C0 (en) | Meniscal repair devices and systems | |
| DK3555814T3 (da) | Udførelse af average pooling i hardware | |
| HUE069411T2 (hu) | Fotovoltaikus rendszer | |
| DK3985986T3 (da) | Systemer og fremgangsmåder til at ændre størrelsen af indhold baseret på indholdets relative betydning | |
| EP3360320C0 (en) | PROJECTION SYSTEMS AND METHODS | |
| EP3483791A4 (en) | CALCULATION SYSTEM USING A HIERARCHICAL NETWORK | |
| DK3606760T3 (da) | Indretning og fremgangsmåde til belægning af overflader | |
| EP3341387A4 (en) | SIALYL TRANSFERASE INHIBITORS AND USES THEREOF | |
| DK3222529T3 (da) | Optimering af rækkevidden af et luftfartøjsdockingsystem | |
| DK3303614T3 (da) | Forbedret anvendelse af overfladeprimere i klynger | |
| DK3137560T3 (da) | Fremgangsmåde til belægning af overflader i kemiske installationer | |
| DK3481970T3 (da) | Termokemisk forarbejdning af eksoterme metalsystemer | |
| DK3188051T3 (da) | Systemer og fremgangsmåder til generering af søgeskabeloner | |
| DK3265555T3 (da) | T-celle-ekspansion | |
| EP3423626A4 (en) | Electroconductive coating | |
| DK3538960T3 (da) | Sekvens styring af programmoduler | |
| IT201700028116A1 (it) | Gruppo di attuazione rotativo-lineare | |
| DK3729441T3 (da) | Påvisning af mikrosatellit instabilitet | |
| EP3333979A4 (en) | SHEET-TYPE METAMATERIAL AND SHEET-LIKE LENS |