Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing
Machine learning, particularly in the form of deep learning (DL), has driven most of the recent fundamental developments in artificial intelligence (AI). DL is based on computational models that are, to a certain extent, bio‐inspired, as they rely on networks of connected simple computing units oper...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-11-01
|
| Series: | Advanced Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/aisy.202000085 |
| _version_ | 1828837615864905728 |
|---|---|
| author | Adnan Mehonic Abu Sebastian Bipin Rajendran Osvaldo Simeone Eleni Vasilaki Anthony J. Kenyon |
| author_facet | Adnan Mehonic Abu Sebastian Bipin Rajendran Osvaldo Simeone Eleni Vasilaki Anthony J. Kenyon |
| author_sort | Adnan Mehonic |
| collection | DOAJ |
| description | Machine learning, particularly in the form of deep learning (DL), has driven most of the recent fundamental developments in artificial intelligence (AI). DL is based on computational models that are, to a certain extent, bio‐inspired, as they rely on networks of connected simple computing units operating in parallel. The success of DL is supported by three factors: availability of vast amounts of data, continuous growth in computing power, and algorithmic innovations. The approaching demise of Moore's law, and the consequent expected modest improvements in computing power that can be achieved by scaling, raises the question of whether the progress will be slowed or halted due to hardware limitations. This article reviews the case for a novel beyond‐complementary metal–oxide–semiconductor (CMOS) technology—memristors—as a potential solution for the implementation of power‐efficient in‐memory computing, DL accelerators, and spiking neural networks. Central themes are the reliance on non‐von‐Neumann computing architectures and the need for developing tailored learning and inference algorithms. To argue that lessons from biology can be useful in providing directions for further progress in AI, an example‐based reservoir computing is briefly discussed. At the end, speculation is given on the “big picture” view of future neuromorphic and brain‐inspired computing systems. |
| first_indexed | 2024-12-12T18:45:44Z |
| format | Article |
| id | doaj.art-ebb5ac157c1c4f68ab829098c26feabf |
| institution | Directory Open Access Journal |
| issn | 2640-4567 |
| language | English |
| last_indexed | 2024-12-12T18:45:44Z |
| publishDate | 2020-11-01 |
| publisher | Wiley |
| record_format | Article |
| series | Advanced Intelligent Systems |
| spelling | doaj.art-ebb5ac157c1c4f68ab829098c26feabf2022-12-22T00:15:32ZengWileyAdvanced Intelligent Systems2640-45672020-11-01211n/an/a10.1002/aisy.202000085Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired ComputingAdnan Mehonic0Abu Sebastian1Bipin Rajendran2Osvaldo Simeone3Eleni Vasilaki4Anthony J. Kenyon5Department of Electronic & Electrical Engineering UCL Torrington Place London WC1E 7JE UKIBM Research Europe Rüschlikon Zurich 8803 SwitzerlandDepartment of Engineering King's College London London WC2R 2LS UKDepartment of Engineering King's College London London WC2R 2LS UKDepartment of Computer Science University of Sheffield Regent Court (DCS) 211 Portobello Sheffield S1 4DP South Yorkshire UKDepartment of Electronic & Electrical Engineering UCL Torrington Place London WC1E 7JE UKMachine learning, particularly in the form of deep learning (DL), has driven most of the recent fundamental developments in artificial intelligence (AI). DL is based on computational models that are, to a certain extent, bio‐inspired, as they rely on networks of connected simple computing units operating in parallel. The success of DL is supported by three factors: availability of vast amounts of data, continuous growth in computing power, and algorithmic innovations. The approaching demise of Moore's law, and the consequent expected modest improvements in computing power that can be achieved by scaling, raises the question of whether the progress will be slowed or halted due to hardware limitations. This article reviews the case for a novel beyond‐complementary metal–oxide–semiconductor (CMOS) technology—memristors—as a potential solution for the implementation of power‐efficient in‐memory computing, DL accelerators, and spiking neural networks. Central themes are the reliance on non‐von‐Neumann computing architectures and the need for developing tailored learning and inference algorithms. To argue that lessons from biology can be useful in providing directions for further progress in AI, an example‐based reservoir computing is briefly discussed. At the end, speculation is given on the “big picture” view of future neuromorphic and brain‐inspired computing systems.https://doi.org/10.1002/aisy.202000085deep learningin-memory computingmemristorsneuromorphic systemspower-efficient artificial intelligencespiking neural networks |
| spellingShingle | Adnan Mehonic Abu Sebastian Bipin Rajendran Osvaldo Simeone Eleni Vasilaki Anthony J. Kenyon Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing Advanced Intelligent Systems deep learning in-memory computing memristors neuromorphic systems power-efficient artificial intelligence spiking neural networks |
| title | Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing |
| title_full | Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing |
| title_fullStr | Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing |
| title_full_unstemmed | Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing |
| title_short | Memristors—From In‐Memory Computing, Deep Learning Acceleration, and Spiking Neural Networks to the Future of Neuromorphic and Bio‐Inspired Computing |
| title_sort | memristors from in memory computing deep learning acceleration and spiking neural networks to the future of neuromorphic and bio inspired computing |
| topic | deep learning in-memory computing memristors neuromorphic systems power-efficient artificial intelligence spiking neural networks |
| url | https://doi.org/10.1002/aisy.202000085 |
| work_keys_str_mv | AT adnanmehonic memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing AT abusebastian memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing AT bipinrajendran memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing AT osvaldosimeone memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing AT elenivasilaki memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing AT anthonyjkenyon memristorsfrominmemorycomputingdeeplearningaccelerationandspikingneuralnetworkstothefutureofneuromorphicandbioinspiredcomputing |