IP Library › Granted Patent US 12,314,866
Granted Patent B2
US 12,314,866 · App. 17/138,334 · Granted May 27, 2025

Parallel processing of network model operations

Inventor: Shaoli Liu (Shanghai, CN)
Assignee: SHANGHAI CAMBRICON INFORMATION TECHNOLOGY CO., LTD
G06N5/02G06F7/57G06F11/3636G06N3/063G06N3/08
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,314,866
App. No.
17/138,334
Granted
May 27, 2025
Kind
B2
Abstract

The present application relates to an operation device and an operation method. The operation device includes a plurality of operation modules. The plurality of operation modules complete an operation of a network model by executing corresponding computational sub-commands in parallel. Each operation module includes at least one operation unit configured to execute a first computational sub-command using first computational sub-data; and a storage unit configured to store the first computational sub-data. The first computational sub-data includes data needed for executing the first computational sub-command. The embodiments of the present application reduces bandwidth requirements for data access and reduces computation and equipment costs.

Claims (27)

1. An operation device comprising a plurality of operation modules configured to execute computational sub-commands in parallel to complete a computational operation of a network model, wherein, each of the operation modules includes:

at least one operation unit configured to execute a first computational sub-command using first computational sub-data, wherein the first computational sub-command of each of the operation modules is part of the computational operation of the network model and the operation modules each execute the first computational sub-command in parallel; and

a separate storage unit configured to store the first computational sub-data, wherein the first computational sub-data includes data for executing the first computational sub-command;

wherein at least two operation modules in the plurality of operation modules have a data connection and a data transfer channel is built between the at least two operation modules based on the data connection for one of the at least two operation modules to obtain data stored in a separate storage unit of another of the at least two operation modules; and

wherein the at least one operation unit in the at least two operation modules is configured to execute the first computational sub-command using the first computational sub-data and second computational sub-data, and wherein the second computational sub-data is stored in the separate storage unit of the other operation module of the at least two operation modules.

2. The operation device of claim 1 , further comprising a control module configured to allocate a corresponding computational sub-command and corresponding computational sub-data to the plurality of operation modules according to a received computational command and computational data, wherein,

the received computational command includes a plurality of computational sub-commands, and the received computational data includes computational sub-data required for executing the plurality of computational sub-commands.

3. The operation device of claim 2 , wherein, the control module includes:

a first allocation unit configured to divide the computational command into a plurality of computational sub-commands according to a computational logic relationship in the computational command, and correspondingly allocate the plurality of computational sub-commands to different operation modules for operation, and

a second allocation unit configured to determine computational sub-data required for executing the computational sub-commands, and allocate the computational sub-data to a corresponding operation module for storage.

4. The operation device of claim 2 , wherein, the control module is further configured to selectively connect operation modules according to a computational logic relationship between the computational sub-commands.

5. The operation device of claim 1 , wherein, the operation unit includes a network processing circuit or a computing circuit.

6. The operation device of claim 1 , wherein, the storage unit includes at least one of a register, a register file, and a memory array.

7. A data processing chip comprising the operation device of claim 1 .

8. An electronic equipment comprising the data processing chip of claim 7 .

9. An operation method, wherein, the method is applied to an operation device, wherein the operation device comprises a plurality of operation modules configured to execute computational sub-commands in parallel to complete a computational operation of a network model, wherein at least two operation modules in the plurality of operation modules are connected by a data connection and a data transfer channel is built between the at least two operation modules based on the data connection, and wherein the method comprises:

storing first computational sub-data using a storage unit of an operation module of the at least two operation modules, wherein the first computational sub-data includes data for executing a first computational sub-command;

storing second computational sub-data using a separate storage unit of another operation module of the at least two operation modules, wherein the second computational sub-data includes data for executing the first computational sub-command; and

executing the first computational sub-command by an operation unit of the operation module using the first computational sub-data and the second computational sub-data, wherein the second computational sub-data is obtained via the data transfer channel from the separate storage unit in the other operation module of the at least two operation modules.

10. The method of claim 9 , further comprising:

allocating a computational sub-command and corresponding computational sub-data to the operation modules according to a received computational command and computational data, wherein,

the received computational command includes a plurality of computational sub-commands, and the received computational data includes computational sub-data required for executing the computational sub-commands.

11. The method of claim 10 , wherein, the allocating the computational sub-command and the corresponding computational sub-data to the operation modules according to the received computational command and computational data includes:

using a first allocation unit to divide the computational command into a plurality of computational sub-commands according to a computational logic relationship in the computational command, and correspondingly allocating the plurality of computational sub-commands to different operation modules for operation, and

using a second allocation unit to determine computational sub-data required for executing the computational sub-commands, and allocating the computational sub-data to a corresponding operation module for storage.

12. The method of claim 10 , further comprising:

selectively connecting operation modules according to a computational logic relationship between the computational sub-commands.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 30, 2020
From: LIU, SHAOLI
To: SHANGHAI CAMBRICON INFORMATION TECHNOLOGY CO., LTD.
Reel/Frame 054779/0823 →
Priority Claims (1)
CN 201810782174.6 · Jul 17, 2018 · national
Continuity (2)
Continuation In Part PCTCN2019092805 · Jun 25, 2019
Related Publication 20210117810A1 · Apr 22, 2021
References Cited (280)
US 4939642A · Blank · 1990 [cited by examiner]
US 5052043A · Gaborski · 1991 [cited by applicant]
US 5361367A · Fijany · 1994 [cited by examiner]
US 5790880A · Ireton · 1998 [cited by examiner]
US 6144977A · Giangarra et al. · 2000 [cited by applicant]
US 6452149B1 · Yamashita · 2002 [cited by examiner]
US 6671796B1 · Sudharsanan et al. · 2003 [cited by applicant]
US 6715065B1 · Ebata et al. · 2004 [cited by applicant]
US 6928535B2 · Yamashita · 2005 [cited by examiner]
US 6931639B1 · Eickemeyer · 2005 [cited by applicant]
US 7236995B2 · Hinds · 2007 [cited by applicant]
US 7242414B1 · Thekkath et al. · 2007 [cited by applicant]
US 7406451B2 · Mrziglod et al. · 2008 [cited by applicant]
US 7721128B2 · Johns et al. · 2010 [cited by applicant]
US 7945607B2 · Hinds · 2011 [cited by applicant]
US 8051117B2 · Lundvall et al. · 2011 [cited by applicant]
US 8106912B2 · Koga · 2012 [cited by examiner]
US 8190664B2 · Lundvall et al. · 2012 [cited by applicant]
US 8560591B2 · Lundvall et al. · 2013 [cited by applicant]
US 8694572B2 · Samy et al. · 2014 [cited by applicant]
US 8730248B2 · Sasaki · 2014 [cited by examiner]
US 8762438B2 · Lundvall et al. · 2014 [cited by applicant]
US 8924455B1 · Barman et al. · 2014 [cited by applicant]
US 8949550B2 · Choi · 2015 [cited by examiner]
US 9412366B2 · Wilensky et al. · 2016 [cited by applicant]
US 9916531B1 · Zivkovic et al. · 2018 [cited by applicant]
US 10187568B1 · Tran et al. · 2019 [cited by applicant]
US 10191881B2 · Goh · 2019 [cited by examiner]
US 10224954B1 · Madduri et al. · 2019 [cited by applicant]
US 10360304B1 · Alvarez et al. · 2019 [cited by applicant]
US 10372507B2 · Venkatesh · 2019 [cited by examiner]
US 10427306B1 · Quinlan et al. · 2019 [cited by applicant]
US 10656942B2 · Madduri et al. · 2020 [cited by applicant]
US 10929744B2 · Li et al. · 2021 [cited by applicant]
US 11568269B2 · Du · 2023 [cited by examiner]
US 20020138714A1 · Leibholz et al. · 2002 [cited by applicant]
US 20030167460A1 · Desai et al. · 2003 [cited by applicant]
US 20040250164A1 · Ahmad et al. · 2004 [cited by applicant]
US 20050138327A1 · Tabei · 2005 [cited by applicant]
US 20060161375A1 · Duberstein et al. · 2006 [cited by applicant]
US 20070220076A1 · Hinds · 2007 [cited by applicant]
US 20070273699A1 · Sasaki · 2007 [cited by examiner]
US 20080148120A1 · Seuring · 2008 [cited by applicant]
US 20090106528A1 · Koga · 2009 [cited by examiner]
US 20090113186A1 · Kato et al. · 2009 [cited by applicant]
US 20090125293A1 · Lefurgy et al. · 2009 [cited by applicant]
US 20100073068A1 · Cho et al. · 2010 [cited by applicant]
US 20110060587A1 · Phillips et al. · 2011 [cited by applicant]
US 20110301777A1 · Cox et al. · 2011 [cited by applicant]
US 20120316845A1 · Grey et al. · 2012 [cited by applicant]
US 20130024621A1 · Choi · 2013 [cited by examiner]
US 20130054110A1 · Sata · 2013 [cited by applicant]
US 20130332610A1 · Beveridge · 2013 [cited by applicant]
US 20140081625A1 · Wilensky et al. · 2014 [cited by applicant]
US 20140164737A1 · Collange et al. · 2014 [cited by applicant]
US 20140249814A1 · Nakano et al. · 2014 [cited by applicant]
US 20150134581A1 · Doeding et al. · 2015 [cited by applicant]
US 20150370303A1 · Krishnaswamy et al. · 2015 [cited by applicant]
US 20160026231A1 · Ignowski et al. · 2016 [cited by applicant]
US 20160054922A1 · Awasthi et al. · 2016 [cited by applicant]
US 20160124710A1 · Lutz et al. · 2016 [cited by applicant]
US 20160170866A1 · Ioualalen et al. · 2016 [cited by applicant]
US 20160328645A1 · Lin et al. · 2016 [cited by applicant]
US 20160328647A1 · Lin et al. · 2016 [cited by applicant]
US 20170061279A1 · Yang et al. · 2017 [cited by applicant]
US 20170090956A1 · Linsky · 2017 [cited by applicant]
US 20170103022A1 · Kreinin et al. · 2017 [cited by applicant]
US 20170142327A1 · Bayani · 2017 [cited by applicant]
US 20170161604A1 · Craddock et al. · 2017 [cited by applicant]
US 20170221176A1 · Munteanu et al. · 2017 [cited by applicant]
US 20170257079A1 · Jain et al. · 2017 [cited by applicant]
US 20170262959A1 · Lee et al. · 2017 [cited by applicant]
US 20170316307A1 · Koster et al. · 2017 [cited by applicant]
US 20170316312A1 · Goyal et al. · 2017 [cited by applicant]
US 20170344882A1 · Ambrose et al. · 2017 [cited by applicant]
US 20170353163A1 · Gazneli et al. · 2017 [cited by applicant]
US 20170357530A1 · Shih et al. · 2017 [cited by applicant]
US 20170357910A1 · Sommer et al. · 2017 [cited by applicant]
US 20180046903A1 · Yao et al. · 2018 [cited by applicant]
US 20180088996A1 · Rossi et al. · 2018 [cited by applicant]
US 20180096243A1 · Patil et al. · 2018 [cited by applicant]
US 20180157464A1 · Lutz et al. · 2018 [cited by applicant]
US 20180288440A1 · Chao · 2018 [cited by applicant]
US 20180293517A1 · Browne et al. · 2018 [cited by applicant]
US 20180300931A1 · Vembu et al. · 2018 [cited by applicant]
US 20180322391A1 · Wu et al. · 2018 [cited by applicant]
US 20180357541A1 · Chen et al. · 2018 [cited by applicant]
US 20180367729A1 · Parasnis et al. · 2018 [cited by applicant]
US 20180373976A1 · Woo · 2018 [cited by applicant]
US 20190026250A1 · Das Sarma · 2019 [cited by examiner]
US 20190034784A1 · Li et al. · 2019 [cited by applicant]
US 20190042925A1 · Choe et al. · 2019 [cited by applicant]
US 20190050710A1 · Wang et al. · 2019 [cited by applicant]
US 20190057696A1 · Ogawa · 2019 [cited by applicant]
US 20190114142A1 · Yoda et al. · 2019 [cited by applicant]
US 20190122094A1 · Chen et al. · 2019 [cited by applicant]
US 20190122119A1 · Husain · 2019 [cited by applicant]
US 20190138372A1 · Tee · 2019 [cited by applicant]
US 20190164285A1 · Nye et al. · 2019 [cited by applicant]
US 20190180170A1 · Huang et al. · 2019 [cited by applicant]
US 20190199370A1 · Madduri et al. · 2019 [cited by applicant]
US 20190205737A1 · Bleiweiss et al. · 2019 [cited by applicant]
US 20190205746A1 · Nurvitadhi et al. · 2019 [cited by applicant]
US 20190220734A1 · Ferdman et al. · 2019 [cited by applicant]
US 20190228762A1 · Wang et al. · 2019 [cited by applicant]
US 20190251429A1 · Du et al. · 2019 [cited by applicant]
US 20190265949A1 · Ito · 2019 [cited by applicant]
US 20190278677A1 · Terechko et al. · 2019 [cited by applicant]
US 20190294968A1 · Vantrease et al. · 2019 [cited by applicant]
US 20190339937A1 · Lo et al. · 2019 [cited by applicant]
US 20200005424A1 · Appu et al. · 2020 [cited by applicant]
US 20200097799A1 · Divakar et al. · 2020 [cited by applicant]
US 20200117453A1 · Zhang et al. · 2020 [cited by applicant]
US 20200117614A1 · Zhang et al. · 2020 [cited by applicant]
US 20200125508A1 · Liu et al. · 2020 [cited by applicant]
US 20200126554A1 · Chen et al. · 2020 [cited by applicant]
US 20200126555A1 · Chen et al. · 2020 [cited by applicant]
US 20200142748A1 · Liu et al. · 2020 [cited by applicant]
US 20200159527A1 · Zhang et al. · 2020 [cited by applicant]
US 20200159530A1 · Zhang et al. · 2020 [cited by applicant]
US 20200159531A1 · Zhang et al. · 2020 [cited by applicant]
US 20200159532A1 · Zhang et al. · 2020 [cited by applicant]
US 20200159533A1 · Zhang et al. · 2020 [cited by applicant]
US 20200159534A1 · Li et al. · 2020 [cited by applicant]
US 20200160162A1 · Zhang et al. · 2020 [cited by applicant]
US 20200160163A1 · Liu et al. · 2020 [cited by applicant]
US 20200160219A1 · Zhang et al. · 2020 [cited by applicant]
US 20200160220A1 · Zhang et al. · 2020 [cited by applicant]
US 20200160221A1 · Zhang et al. · 2020 [cited by applicant]
US 20200160222A1 · Zhang et al. · 2020 [cited by applicant]
US 20200168227A1 · Chen et al. · 2020 [cited by applicant]
US 20200174547A1 · Fang et al. · 2020 [cited by applicant]
US 20200183752A1 · Liu et al. · 2020 [cited by applicant]
US 20200241874A1 · Chen et al. · 2020 [cited by applicant]
US 20200257972A1 · Miniskar et al. · 2020 [cited by applicant]
US 20200334041A1 · Zhang et al. · 2020 [cited by applicant]
US 20200334522A1 · Zhang et al. · 2020 [cited by applicant]
US 20200334572A1 · Zhang et al. · 2020 [cited by applicant]
US 20200387800A1 · Du · 2020 [cited by examiner]
US 20200394522A1 · Liu et al. · 2020 [cited by applicant]
US 20200394523A1 · Liu et al. · 2020 [cited by applicant]
US 20210042889A1 · Pei · 2021 [cited by applicant]
US 20210061028A1 · Da Deppo et al. · 2021 [cited by applicant]
US 20210117768A1 · Liu et al. · 2021 [cited by applicant]
US 20210182177A1 · Su et al. · 2021 [cited by applicant]
US 20210264270A1 · Liu et al. · 2021 [cited by applicant]
US 20210286688A1 · Liu et al. · 2021 [cited by applicant]
US 20210334007A1 · Liu et al. · 2021 [cited by applicant]
US 20210334137A1 · Zhang et al. · 2021 [cited by applicant]
US 20210341989A1 · Chen et al. · 2021 [cited by applicant]
US 20210374510A1 · Liu et al. · 2021 [cited by applicant]
US 20210374511A1 · Liu et al. · 2021 [cited by applicant]
CN 1503858A · 2004 [cited by applicant]
CN 1503958A · 2004 [cited by applicant]
CN 1851668A · 2006 [cited by applicant]
CN 101572829A · 2009 [cited by applicant]
CN 102270042A · 2011 [cited by applicant]
CN 102404673A · 2012 [cited by applicant]
CN 102684701A · 2012 [cited by applicant]
CN 102761509A · 2012 [cited by applicant]
CN 102789413A · 2012 [cited by applicant]
CN 102903089A · 2013 [cited by applicant]
CN 102981854A · 2013 [cited by applicant]
CN 103152673A · 2013 [cited by applicant]
CN 104914977A · 2015 [cited by applicant]
CN 105389158A · 2016 [cited by applicant]
CN 103534664A · 2016 [cited by applicant]
CN 105893419A · 2016 [cited by applicant]
CN 105978611A · 2016 [cited by applicant]
CN 106156310A · 2016 [cited by applicant]
CN 106354568A · 2017 [cited by applicant]
CN 106406812A · 2017 [cited by applicant]
CN 106469291A · 2017 [cited by applicant]
CN 106485316A · 2017 [cited by applicant]
CN 106502626A · 2017 [cited by applicant]
CN 106570559A · 2017 [cited by applicant]
CN 106650922A · 2017 [cited by applicant]
CN 106814639A · 2017 [cited by applicant]
CN 106951587A · 2017 [cited by applicant]
CN 106951962A · 2017 [cited by applicant]
CN 106997236A · 2017 [cited by applicant]
CN 107003988A · 2017 [cited by applicant]
CN 107025629A · 2017 [cited by applicant]
CN 107368174A · 2017 [cited by applicant]
CN 107451654A · 2017 [cited by applicant]
CN 107451658A · 2017 [cited by applicant]
CN 107608715A · 2018 [cited by applicant]
CN 107644254A · 2018 [cited by applicant]
CN 107688855A · 2018 [cited by applicant]
CN 107797913A · 2018 [cited by applicant]
CN 108053028A · 2018 [cited by applicant]
CN 104899641B · 2018 [cited by applicant]
CN 108337000A · 2018 [cited by applicant]
CN 108510067A · 2018 [cited by applicant]
CN 108717570A · 2018 [cited by applicant]
CN 109062540A · 2018 [cited by applicant]
CN 109063820A · 2018 [cited by applicant]
CN 109146057A · 2019 [cited by applicant]
CN 109214509A · 2019 [cited by applicant]
CN 109389219A · 2019 [cited by applicant]
CN 109472353A · 2019 [cited by applicant]
CN 109800877A · 2019 [cited by applicant]
CN 109902745 · 2019 [cited by applicant]
CN 109934331A · 2019 [cited by applicant]
CN 109993296A · 2019 [cited by applicant]
CN 110059733A · 2019 [cited by applicant]
CN 11055450A · 2019 [cited by applicant]
CN 110780845A · 2020 [cited by applicant]
EP 0789296A1 · 1997 [cited by applicant]
EP 2703945A2 · 2014 [cited by applicant]
EP 3106997A2 · 2016 [cited by applicant]
EP 3407268A1 · 2018 [cited by applicant]
EP 3920026A1 · 2021 [cited by examiner]
EP 3992863A1 · 2022 [cited by examiner]
JP H03075860A · 1989 [cited by applicant]
JP H09265379A · 1997 [cited by applicant]
JP 2009134433A · 2012 [cited by applicant]
JP 2013514570A · 2013 [cited by applicant]
JP 2014199464A · 2014 [cited by applicant]
JP 2015176158A · 2015 [cited by applicant]
JP 201826114A · 2018 [cited by applicant]
JP 2019519852A · 2019 [cited by applicant]
WO WO9725670A1 · 1997 [cited by examiner]
WO 2008153194A1 · 2008 [cited by applicant]
WO 2016186823A1 · 2016 [cited by applicant]
WO 2017138220A1 · 2017 [cited by applicant]
WO 2017185412A1 · 2017 [cited by applicant]
WO WO2023045638A1 · 2023 [cited by examiner]
‘Placement and Interconnection of Systolic Processing Elements: A New Lu Decomposition Algorithm’ by Marina C. Chen, Oct. 1986. (Year: 1986). [cited by examiner]
‘Engineering of Massively Interconnected Computer Systems’ by Michael Jenkins et al., Underwater Systems Department, Oct. 1, 1991. (Year: 1991). [cited by examiner]
‘Comparison of Internal and External Watchdog Timers’ by: Eric Schlaepfer, Applications Engineer, Jun. 5, 2008. (Year: 2008). [cited by examiner]
Chen et al., “DianNao: A Small-Footprint High-Throughput Accelerator for Ubiquitous Machine-Learning,” ASPLOS '14 Proceedings of the 19th international conference on Architectural support for programming languages and o… [cited by applicant]
Chen et al., “A Small-Footprint Accelerator for Large-Scale Neural Networks,” ACM Transactions on Computer Systems (TOCS), vol. 33, Issue 2, May 1, 2015, Article No. 6, ACM, pp. 1-27, New York, NY, USA; available: https… [cited by applicant]
Chen, Y., et al., “DaDianNao: A Machine-Learning Supercomputer,” MICRO-47 Proceedings of the 47th Annual IEEE/ACM International Symposium on Microarchitecture, pp. 609-622, Cambridge, United Kingdom—Dec. 13, 2014; avail… [cited by applicant]
Luo, T., et al., “DaDianNao: A Neural Network Supercomputer,” Published in: IEEE Transactions on Computers, vol. 66 , Issue: 1, pp. 73-88, Date of Publication: May 30, 2016; available: https://ieeexplore.ieee.org/docume… [cited by applicant]
Liu, D., et al., “PuDianNao: A Polyvalent Machine Learning Accelerator,” ASPLOS '15 Proceedings of the Twentieth International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 369… [cited by applicant]
Du, Z., et al., “ShiDianNao: Shifting Vision Processing Closer to the Sensor,” ISCA '15 Proceedings of the 42nd Annual International Symposium on Computer Architecture, pp. 92-104, Portland, Oregon—Jun. 13, 2015; availa… [cited by applicant]
Du, Z., et al., “An Accelerator for High Efficient Vision Processing,” Published in: IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 36 , Issue: 2, Jun. 22, 2016, pp. 227-240; availab… [cited by applicant]
Liu, S., et al., “Cambricon: An Instruction Set Architecture for Neural Networks,” 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA), Seoul, South Korea, Jun. 18, 2016; pp. 393-405, avail… [cited by applicant]
Zhang, S. et al., “Cambricon-X: An Accelerator for Sparse Neural Networks,” Published in: 2016 49th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO), Taipei, Taiwan, Oct. 15, 2016; 12 pages, availabl… [cited by applicant]
Chen, Y., et al., “DianNao Family: Energy-Efficient Hardware Accelerators for Machine Learning,” Communications of the ACM, vol. 59 Issue 11, Oct. 28, 2016, pp. 105-112, ACM New York, NY, USA; available: https://dl.acm.… [cited by applicant]
Vivienne Sze et al., “Efficient Processing of Deep Neural Networks: A Tutorial and Survey”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853, Mar. 27, 2017 (Mar. 27, 2017), pp.… [cited by applicant]
Liu Shaoli et al., “Cambricon: An Instruction Set Architecture for Neural Networks”, 2013 21st International Conference on Program Comprehension (ICPC); [International Symposium on Computer Architecture.(ISCA)], IEEE, U… [cited by applicant]
IBM, “PowerPC Microprocessor Family: Vector/SIMD Multimedia Extension Technology Programming Environments Manual Programming Environments Manual”, Aug. 22, 2005 (Aug. 22, 2005), pp. 170-171, XP055673033, Retrieved from … [cited by applicant]
Sumod Mohan, “Accuracy and Multi-Core Performance of Machine Learning Algorithms for Handwritten Character Recognition”, Aug. 1, 2009 (Aug. 1, 2009), 60 pages, XP055673941, Retrieved from the Internet: URL:https://tiger… [cited by applicant]
Anonymous, “Control unit—Wikipedia”, May 8, 2017 (May 8, 2017), 9 pages, XP055673879, Retrieved from the Internet: URL: https://web.archive.org/web/20170508110348/https://en.wikipedia.org/ wiki/Control unit [retrieved o… [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212749.6 dated Mar. 18, 2020, 6 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212750.4 dated Mar. 18, 2020, 5 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212751.2 dated Mar. 18, 2020, 7 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212752.0 dated Mar. 18, 2020, 6 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19214004.4 dated Mar. 18, 2020, 5 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19213389.0 dated Mar. 18, 2020, 5 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212753.8 dated Mar. 18, 2020, 7 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212754.6 dated Mar. 18, 2020, 6 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212755.3 dated Mar. 18, 2020, 6 pages. [cited by applicant]
Joel Emer et al., “DNN Accelerator Architectures”, CICS/MTL Tutorial, Jan. 1, 2017 (Jan. 1, 2017), XP055672691, 74 pages. Retrieved from the Internet: URL:https://www.rle.mit.edu/eems/wp-content/uploads/2017/031Tutorial… [cited by applicant]
Chen Yu-Hsin et al., “Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural getworks”, IEEE Journal of Solid-State Circuits, IEEE, USA, vol. 52, No. 1, Jan. 1, 2017 (Jan. 1, 2017), pp. 12… [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212756.1 dated Mar. 18, 2020, 7 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 18906652.6 dated Mar. 18, 2020, 5 pages. [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19212746.2 dated Mar. 18, 2020, 8 pages. [cited by applicant]
Kalathingal Sajith et al., “Dynamic Inter-Thread Vectorization Architecture: Extracting OLP from TLP”, 2016 28th International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), IEEE, Oct. 26,… [cited by applicant]
Li et al., “Using Artificial Neural Network for Predicting Thread Partitioning in Speculative Multithreading”, IEEE, 2015, pp. 823-826. [cited by applicant]
Na et al., “Speeding up Convolutional Neural Network Training with Dynamic Precision Scaling and Flexible MultiplierAccumulator”, Section 2 Proposed Approach: Concept, ACM, Aug. 8-10, 2016, 6 pages. [cited by applicant]
Hanlon, Jamie, “Why is so much memory needed for deep neural networks?”, URL: https://www.graphcore.ai/posts/why-is-so-much-memory-needed-for-deep-neural-networks, Jan. 31, 2017, 6 pages. [cited by applicant]
Extended European Search Report for Application No. 19215861.6 mailed May 15, 2020, 17 pages. [cited by applicant]
Extended European Search Report for Application No. 19215862.4 mailed May 15, 2020, 17 pages. [cited by applicant]
Sumina Yamashita, et al., “A Method to create illustrate images using DCGAN,” JISJ SIG Technical Report, vol. 2017-MPS-112 No. 16, Feb. 27, 2017, 8 pages. (Translation of Abstract Included). [cited by applicant]
European Patent Office, Extended European Search Report for European Application No. 19218382.0 dated Apr. 24, 2020, 11 pages. [cited by applicant]
Yi Yang et al., “Deploy Large-Scale Deep Neural Networks in Resource Constrained Io T Devices with Local Quantization Region”, arxiv.org, Cornell University Library, 201 Olin Library Cornell University Ithaca, NY 14853,… [cited by applicant]
Gysel Philipp et al., “Ristretto: A Framework for Empirical Study of Resource-Efficient Inference in Convolutional Neural Networks”, IEEE Transactions on Neural Networks and Learning Systems, IEEE, Piscataway, NJ, USA, … [cited by applicant]
Hsu Jeremy, “For sale: deep learning [News]”, IEEE Spectrum, IEEE Inc. New York, US, vol. 53, No. 8, Aug. 1, 2016 (Aug. 1, 2016), pp. 12-13, XP011620787, ISSN: 0018-9235, DOI: 10.1109/MSPEC.2016.7524158 [retrieved on Ju… [cited by applicant]
Song Mingcong et al., “In-Situ AI: Towards Autonomous and Incremental Deep Learning for IoT Systems”, 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA), IEEE, Feb. 24, 2018 (Feb. 24, 201… [cited by applicant]
Kallam Suresh et al., “Evaluating the Performance of Deep Learning Techniques on Classification Using Tensor Flow Application”, 2018 International Conference on Advances in Computing and Communication Engineering (ICACC… [cited by applicant]
Olariu Cristian et al., “A Cloud-Based AI Framework for Machine Learning Orchestration: A “Driving or Not-Driving” Case-Study for Self-Driving Cars”, 2019 IEEE Intelligent Vehicles Symposium (IV). IEEE, Jun. 9, 2019 (Ju… [cited by applicant]
European Patent Office, extended European search report for Application No. 19216754.2 mailed May 8, 2020, 8 pages. [cited by applicant]
Extended European Search Report for EP Application No. 19214324.6 mailed Oct. 1, 2020, 11 pages. [cited by applicant]
International Searching Authority, International Search Report for PCT Application No. PCT/CN2019/093144 mailed Oct. 9, 2019, 6 pages. [cited by applicant]
Communication pursuant to Article 94(3) EPC received for European Application No. 18861574.4, dated Jun. 2, 2022, 20 pages. [cited by applicant]
Pedro O. Domingos; An Efficient and Scalable Architecture for Neural Networks With Backpropagation Learning; 0-7803-9362 2005 IEEE; Jul. 5, 2005; 6 Pages. [cited by applicant]
Pedro 0. Domingos, et al. “An Efficient and Scalable Architecture for Neural Networks With Backpropagation Learning” Proceedings/ 2005 International Conference on Field Programmable Logic and Applications (FPL): Tampere… [cited by applicant]