GB2587021B - Physical implementation of artificial neural networks - Google Patents

Physical implementation of artificial neural networks Download PDF

Info

Publication number
GB2587021B
GB2587021B GB1913275.2A GB201913275A GB2587021B GB 2587021 B GB2587021 B GB 2587021B GB 201913275 A GB201913275 A GB 201913275A GB 2587021 B GB2587021 B GB 2587021B
Authority
GB
United Kingdom
Prior art keywords
neural networks
artificial neural
physical implementation
implementation
physical
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
GB1913275.2A
Other versions
GB2587021A (en
GB201913275D0 (en
Inventor
Mehonic Adnan
Joksas Dovydas
J Kenyon Anthony
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
UCL Business Ltd
Original Assignee
UCL Business Ltd
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 UCL Business Ltd filed Critical UCL Business Ltd
Priority to GB1913275.2A priority Critical patent/GB2587021B/en
Publication of GB201913275D0 publication Critical patent/GB201913275D0/en
Priority to EP20772380.0A priority patent/EP4028955A1/en
Priority to PCT/GB2020/052166 priority patent/WO2021048542A1/en
Publication of GB2587021A publication Critical patent/GB2587021A/en
Application granted granted Critical
Publication of GB2587021B publication Critical patent/GB2587021B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06N3/06—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
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Neurology (AREA)
  • Logic Circuits (AREA)
  • Semiconductor Memories (AREA)
GB1913275.2A 2019-09-13 2019-09-13 Physical implementation of artificial neural networks Active GB2587021B (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
GB1913275.2A GB2587021B (en) 2019-09-13 2019-09-13 Physical implementation of artificial neural networks
EP20772380.0A EP4028955A1 (en) 2019-09-13 2020-09-09 Physical implementation of artificial neural networks
PCT/GB2020/052166 WO2021048542A1 (en) 2019-09-13 2020-09-09 Physical implementation of artificial neural networks

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GB1913275.2A GB2587021B (en) 2019-09-13 2019-09-13 Physical implementation of artificial neural networks

Publications (3)

Publication Number Publication Date
GB201913275D0 GB201913275D0 (en) 2019-10-30
GB2587021A GB2587021A (en) 2021-03-17
GB2587021B true GB2587021B (en) 2023-09-13

Family

ID=68315241

Family Applications (1)

Application Number Title Priority Date Filing Date
GB1913275.2A Active GB2587021B (en) 2019-09-13 2019-09-13 Physical implementation of artificial neural networks

Country Status (3)

Country Link
EP (1) EP4028955A1 (en)
GB (1) GB2587021B (en)
WO (1) WO2021048542A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112990444B (en) * 2021-05-13 2021-09-24 电子科技大学 Hybrid neural network training method, system, equipment and storage medium

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
AGRAWAL et al., "X-CHANGR: Changing Memristive Crossbar Mapping for Mitigating Line-Resistance Induced Accuracy Degradation in Deep Neural Networks", 26 June 2019, available at https://arxiv.org/abs/1907.00285 *
JOKSAS et al., "Committee Machines - A Universal Method to Deal with Non-Idealities in RRAM-Based Neural Networks", 14 September 2019, available at https://arxiv.org/abs/1909.06658 *
LI et al., "2015 52nd ACM/EDAC/IEEE Design Automation Conference (DAC", 2015, IEEE, "Merging the Interface: Power, Area and Accuracy Co-optimization for RRAM Crossbar-based Mixed-Signal Computing System" *
MEHONIC et al., Frontiers in Neuroscience, 2019, volume 13 page 593, "Simulation of Inference Accuracy Using Realistic RRAM Devices" *
PAYVAND et al., "A neuromorphic systems approach to in-memory computing with non-ideal memristive devices: From mitigation to exploitation", 13 July 2018, available at https://arxiv.org/abs/1807.05128 *

Also Published As

Publication number Publication date
WO2021048542A1 (en) 2021-03-18
GB2587021A (en) 2021-03-17
GB201913275D0 (en) 2019-10-30
EP4028955A1 (en) 2022-07-20

Similar Documents

Publication Publication Date Title
GB2581731B (en) Training of artificial neural networks
IL288055A (en) Training of artificial neural networks
EP3794516A4 (en) Gradient adversarial training of neural networks
IL279533A (en) Artificial intelligence-based sequencing
PL3542316T3 (en) Attention-based sequence transduction neural networks
EP3724819A4 (en) Evolutionary architectures for evolution of deep neural networks
GB201611857D0 (en) An artificial neural network
EP3874413A4 (en) Automated generation of neural networks
EP3942483A4 (en) Multiobjective coevolution of deep neural network architectures
CA211530S (en) Humeral implant
CA194246S (en) Humeral implant
EP4099984A4 (en) Artificial synapses
EP4026940A4 (en) Artificial intelligence dryer
EP4172891A4 (en) Artificial intelligence for next best action
EP3794514C0 (en) Switched artificial synapse
GB201804451D0 (en) Artificial neural networks
GB201803083D0 (en) Artificial Neural Networks
GB201907476D0 (en) Multilayer set of neural networks
GB2593093B (en) Artificial intelligence mobile integration
GB201717885D0 (en) Prothesis and method of manufacture
GB2587021B (en) Physical implementation of artificial neural networks
EP3563305A4 (en) Artificial neural network
GB2570018B8 (en) Prosthesis
HUP1900359A1 (en) Artificial neural network
EP4104103A4 (en) Multiple-task neural networks