A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems
Data-intensive computing applications, such as object recognition, time series prediction, and optimization tasks, are becoming increasingly important in several fields, including smart mobility, health, and industry. Because of the large amount of data involved in the computation, the conventional...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Journal on Exploratory Solid-State Computational Devices and Circuits |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9086758/ |
| _version_ | 1818419593096462336 |
|---|---|
| author | Giacomo Pedretti Piergiulio Mannocci Shahin Hashemkhani Valerio Milo Octavian Melnic Elisabetta Chicca Daniele Ielmini |
| author_facet | Giacomo Pedretti Piergiulio Mannocci Shahin Hashemkhani Valerio Milo Octavian Melnic Elisabetta Chicca Daniele Ielmini |
| author_sort | Giacomo Pedretti |
| collection | DOAJ |
| description | Data-intensive computing applications, such as object recognition, time series prediction, and optimization tasks, are becoming increasingly important in several fields, including smart mobility, health, and industry. Because of the large amount of data involved in the computation, the conventional von Neumann architecture suffers from excessive latency and energy consumption due to the memory bottleneck. A more efficient approach consists of in-memory computing (IMC), where computational operations are directly carried out within the data. IMC can take advantage of the rich physics of memory devices, such as their ability to store analog values to be used in matrix-vector multiplication (MVM) and their stochasticity that is highly valuable in the frame of optimization and constraint satisfaction problems (CSPs). This article presents a stochastic spiking neuron based on a phase-change memory (PCM) device for the solution of CSPs within a Hopfield recurrent neural network (RNN). In the RNN, the PCM cell is used as the integrating element of a stochastic neuron, supporting the solution of a typical CSP, namely a Sudoku puzzle in hardware. Finally, the ability to solve Sudoku puzzles using RNNs with PCM-based neurons is studied for increasing size of Sudoku puzzles by a compact simulation model, thus supporting our PCM-based RNN for data-intensive computing. |
| first_indexed | 2024-12-14T12:41:02Z |
| format | Article |
| id | doaj.art-42bba176a9fa44adaf0515249725abfd |
| institution | Directory Open Access Journal |
| issn | 2329-9231 |
| language | English |
| last_indexed | 2024-12-14T12:41:02Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal on Exploratory Solid-State Computational Devices and Circuits |
| spelling | doaj.art-42bba176a9fa44adaf0515249725abfd2022-12-21T23:00:54ZengIEEEIEEE Journal on Exploratory Solid-State Computational Devices and Circuits2329-92312020-01-0161899710.1109/JXCDC.2020.29926919086758A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction ProblemsGiacomo Pedretti0https://orcid.org/0000-0002-4501-8672Piergiulio Mannocci1https://orcid.org/0000-0002-0083-5804Shahin Hashemkhani2https://orcid.org/0000-0003-3629-6424Valerio Milo3https://orcid.org/0000-0002-3305-2742Octavian Melnic4https://orcid.org/0000-0001-9192-3254Elisabetta Chicca5Daniele Ielmini6https://orcid.org/0000-0002-1853-1614Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyFaculty of Technology and the Center of Cognitive Interaction Technology (CITEC), Bielefeld University, Bielefeld, GermanyDipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano and IU.NET, Milan, ItalyData-intensive computing applications, such as object recognition, time series prediction, and optimization tasks, are becoming increasingly important in several fields, including smart mobility, health, and industry. Because of the large amount of data involved in the computation, the conventional von Neumann architecture suffers from excessive latency and energy consumption due to the memory bottleneck. A more efficient approach consists of in-memory computing (IMC), where computational operations are directly carried out within the data. IMC can take advantage of the rich physics of memory devices, such as their ability to store analog values to be used in matrix-vector multiplication (MVM) and their stochasticity that is highly valuable in the frame of optimization and constraint satisfaction problems (CSPs). This article presents a stochastic spiking neuron based on a phase-change memory (PCM) device for the solution of CSPs within a Hopfield recurrent neural network (RNN). In the RNN, the PCM cell is used as the integrating element of a stochastic neuron, supporting the solution of a typical CSP, namely a Sudoku puzzle in hardware. Finally, the ability to solve Sudoku puzzles using RNNs with PCM-based neurons is studied for increasing size of Sudoku puzzles by a compact simulation model, thus supporting our PCM-based RNN for data-intensive computing.https://ieeexplore.ieee.org/document/9086758/Phase change memory (PCM)artificial synapseshopfield neural networkstochastic processoptimization |
| spellingShingle | Giacomo Pedretti Piergiulio Mannocci Shahin Hashemkhani Valerio Milo Octavian Melnic Elisabetta Chicca Daniele Ielmini A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems IEEE Journal on Exploratory Solid-State Computational Devices and Circuits Phase change memory (PCM) artificial synapses hopfield neural network stochastic process optimization |
| title | A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems |
| title_full | A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems |
| title_fullStr | A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems |
| title_full_unstemmed | A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems |
| title_short | A Spiking Recurrent Neural Network With Phase-Change Memory Neurons and Synapses for the Accelerated Solution of Constraint Satisfaction Problems |
| title_sort | spiking recurrent neural network with phase change memory neurons and synapses for the accelerated solution of constraint satisfaction problems |
| topic | Phase change memory (PCM) artificial synapses hopfield neural network stochastic process optimization |
| url | https://ieeexplore.ieee.org/document/9086758/ |
| work_keys_str_mv | AT giacomopedretti aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT piergiuliomannocci aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT shahinhashemkhani aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT valeriomilo aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT octavianmelnic aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT elisabettachicca aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT danieleielmini aspikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT giacomopedretti spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT piergiuliomannocci spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT shahinhashemkhani spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT valeriomilo spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT octavianmelnic spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT elisabettachicca spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems AT danieleielmini spikingrecurrentneuralnetworkwithphasechangememoryneuronsandsynapsesfortheacceleratedsolutionofconstraintsatisfactionproblems |