Simulation platform for pattern recognition based on reservoir computing with memristor networks
Abstract Memristive systems and devices are potentially available for implementing reservoir computing (RC) systems applied to pattern recognition. However, the computational ability of memristive RC systems depends on intertwined factors such as system architectures and physical properties of memri...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-06-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-022-13687-z |
| _version_ | 1811241129714122752 |
|---|---|
| author | Gouhei Tanaka Ryosho Nakane |
| author_facet | Gouhei Tanaka Ryosho Nakane |
| author_sort | Gouhei Tanaka |
| collection | DOAJ |
| description | Abstract Memristive systems and devices are potentially available for implementing reservoir computing (RC) systems applied to pattern recognition. However, the computational ability of memristive RC systems depends on intertwined factors such as system architectures and physical properties of memristive elements, which complicates identifying the key factor for system performance. Here we develop a simulation platform for RC with memristor device networks, which enables testing different system designs for performance improvement. Numerical simulations show that the memristor-network-based RC systems can yield high computational performance comparable to that of state-of-the-art methods in three time series classification tasks. We demonstrate that the excellent and robust computation under device-to-device variability can be achieved by appropriately setting network structures, nonlinearity of memristors, and pre/post-processing, which increases the potential for reliable computation with unreliable component devices. Our results contribute to an establishment of a design guide for memristive reservoirs toward the realization of energy-efficient machine learning hardware. |
| first_indexed | 2024-04-12T13:30:59Z |
| format | Article |
| id | doaj.art-46475ed1ab2c49678fc1619ffb8358cc |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-12T13:30:59Z |
| publishDate | 2022-06-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-46475ed1ab2c49678fc1619ffb8358cc2022-12-22T03:31:10ZengNature PortfolioScientific Reports2045-23222022-06-0112111310.1038/s41598-022-13687-zSimulation platform for pattern recognition based on reservoir computing with memristor networksGouhei Tanaka0Ryosho Nakane1International Research Center for Neurointelligence, The University of TokyoDepartment of Electrical Engineering and Information Systems, Graduate School of Engineering, The University of TokyoAbstract Memristive systems and devices are potentially available for implementing reservoir computing (RC) systems applied to pattern recognition. However, the computational ability of memristive RC systems depends on intertwined factors such as system architectures and physical properties of memristive elements, which complicates identifying the key factor for system performance. Here we develop a simulation platform for RC with memristor device networks, which enables testing different system designs for performance improvement. Numerical simulations show that the memristor-network-based RC systems can yield high computational performance comparable to that of state-of-the-art methods in three time series classification tasks. We demonstrate that the excellent and robust computation under device-to-device variability can be achieved by appropriately setting network structures, nonlinearity of memristors, and pre/post-processing, which increases the potential for reliable computation with unreliable component devices. Our results contribute to an establishment of a design guide for memristive reservoirs toward the realization of energy-efficient machine learning hardware.https://doi.org/10.1038/s41598-022-13687-z |
| spellingShingle | Gouhei Tanaka Ryosho Nakane Simulation platform for pattern recognition based on reservoir computing with memristor networks Scientific Reports |
| title | Simulation platform for pattern recognition based on reservoir computing with memristor networks |
| title_full | Simulation platform for pattern recognition based on reservoir computing with memristor networks |
| title_fullStr | Simulation platform for pattern recognition based on reservoir computing with memristor networks |
| title_full_unstemmed | Simulation platform for pattern recognition based on reservoir computing with memristor networks |
| title_short | Simulation platform for pattern recognition based on reservoir computing with memristor networks |
| title_sort | simulation platform for pattern recognition based on reservoir computing with memristor networks |
| url | https://doi.org/10.1038/s41598-022-13687-z |
| work_keys_str_mv | AT gouheitanaka simulationplatformforpatternrecognitionbasedonreservoircomputingwithmemristornetworks AT ryoshonakane simulationplatformforpatternrecognitionbasedonreservoircomputingwithmemristornetworks |