Yoshihiko Horio
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Yoshihiko Horio has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Takemori Orima
ACRO, Teikyo University
9 shared publications - Shigeo Sato
Research Institute of Electrical Communication, Tohoku University, Sendai 980-8577, Japan
5 shared publications - Satoshi Moriya
Tohoku University
4 shared publications - Hideaki Yamamoto
Tohoku University
3 shared publications - Jordi Madrenas
Department of Electronic Engineering, Universitat Politècnica de Catalunya, Barcelona, Spain
3 shared publications - Mitsumasa Koyanagi
New Industry Creation Hatchery Center (NICHe), Tohoku University, Sendai, Japan
3 shared publications
Latest Publications
- 15 September 2026
- Published by Springer Nature
- p. 307-317
Temporal Evolution of Synaptic Weights in Spatiotemporal Learning Spiking Neural Networks
- 15 September 2026
- Published by Springer Nature
- p. 307-317
Temporal Evolution of Synaptic Weights in Spatiotemporal Learning Spiking Neural Networks
- 1 January 2026
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 17 (3) , 1213-1224
Effect of time history in coincidence among inputs on spatiotemporal learning rule
- 1 January 2026
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 17 (3) , 1213-1224
Effect of time history in coincidence among inputs on spatiotemporal learning rule
- Open Access
- 24 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 72 (10) , 5571-5582
Analog VLSI Implementation of Subthreshold Spiking Neural Networks and Its Application to Reservoir Computing
Neuromorphic computing achieves highly energy-efficient computations while adapting to environmental changes. Processing time series data by spiking neural networks can further reduce the power consumption of neuromorphic computing devices because most of the energy in the network is consumed only when the neuron generates and transmits a spike. In this study, we designed fully analog two-variable spiking neuron and spiking neural network circuits, taking advantage of the physical properties of transistors as analog devices. The energy consumption of the circuit for generating a spike was 22.7 fJ/spike when the MOS transistors were operating in the subthreshold region. The proposed circuits were implemented on an analog very large-scale integrated (VLSI) circuit chip in a 0.18 µm CMOS process. The circuits exhibit complex spike dynamics even under subthreshold operation according to chip measurements. We demonstrated that the spike sequence generated by the spiking neural network circuit was successfully applied to spoken digit recognition tasks via a reservoir computing framework with 14.4 fJ/SOP efficiency. These results provide important insights into edge AI applications of SNN-based neuromorphic hardware.
- 24 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems I: Regular Papers
- Vol. 72 (10) , 5571-5582
Open AccessAnalog VLSI Implementation of Subthreshold Spiking Neural Networks and Its Application to Reservoir Computing
Neuromorphic computing achieves highly energy-efficient computations while adapting to environmental changes. Processing time series data by spiking neural networks can further reduce the power consumption of neuromorphic computing devices because most of the energy in the network is consumed only when the neuron generates and transmits a spike. In this study, we designed fully analog two-variable spiking neuron and spiking neural network circuits, taking advantage of the physical properties of transistors as analog devices. The energy consumption of the circuit for generating a spike was 22.7 fJ/spike when the MOS transistors were operating in the subthreshold region. The proposed circuits were implemented on an analog very large-scale integrated (VLSI) circuit chip in a 0.18 µm CMOS process. The circuits exhibit complex spike dynamics even under subthreshold operation according to chip measurements. We demonstrated that the spike sequence generated by the spiking neural network circuit was successfully applied to spoken digit recognition tasks via a reservoir computing framework with 14.4 fJ/SOP efficiency. These results provide important insights into edge AI applications of SNN-based neuromorphic hardware.
- Open Access
- 1 January 2025
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 16 (3) , 312-334
Brainmorphic computing: Paradigm and hardware
A brainmorphic computing (BMC) paradigm, a new framework for brain-like computing, is necessary for realizing a highly efficient and high-performance brain-specific information processing that emerged evolutionarily in the brain. The BMC paradigm should consider the tight relationship between complex structure and computation, high-dimensional integrated dynamics, embodiment, and computation through biophysical phenomena. Furthermore, a dedicated hardware, i.e., brainmorphic computing hardware (BMC-HW), is mandatory for reproducing the BMC paradigm, through which highly efficient, high-performance, and brain-like computation can be realized as a practical engineering system. It is natural and essential to employ continuous-value and continuous-time physical phenomena in BMC-HW like the brain. This study reviews the BMC paradigm and BMC-HW framework with some implementation examples, which we have developed. In addition, extensions of the BMC paradigm and BMC-HW framework to a complex spatiotemporal information processing through event-driven, continuous-value, continuous-time, asynchronous spiking neural networks are discussed, which is closer to the way the brain processes information. Furthermore, novel device and circuit technologies to realize the extended BMC paradigm are introduced.
- 1 January 2025
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 16 (3) , 312-334
Open AccessBrainmorphic computing: Paradigm and hardware
A brainmorphic computing (BMC) paradigm, a new framework for brain-like computing, is necessary for realizing a highly efficient and high-performance brain-specific information processing that emerged evolutionarily in the brain. The BMC paradigm should consider the tight relationship between complex structure and computation, high-dimensional integrated dynamics, embodiment, and computation through biophysical phenomena. Furthermore, a dedicated hardware, i.e., brainmorphic computing hardware (BMC-HW), is mandatory for reproducing the BMC paradigm, through which highly efficient, high-performance, and brain-like computation can be realized as a practical engineering system. It is natural and essential to employ continuous-value and continuous-time physical phenomena in BMC-HW like the brain. This study reviews the BMC paradigm and BMC-HW framework with some implementation examples, which we have developed. In addition, extensions of the BMC paradigm and BMC-HW framework to a complex spatiotemporal information processing through event-driven, continuous-value, continuous-time, asynchronous spiking neural networks are discussed, which is closer to the way the brain processes information. Furthermore, novel device and circuit technologies to realize the extended BMC paradigm are introduced.
- Open Access
- 1 January 2024
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 15 (4) , 796-810
Spatiotemporal contextual learning network with excitatory and inhibitory synapses for spiking neural network hardware
A spatiotemporal contextual learning network (STCLN) model is suitable for edge applications. To implement the STCLN model as a small analog asynchronous integrated circuit with low power consumption, this study proposes to introduce the excitatory and inhibitory synapses into the STCLN model which can be implemented by an event-driven spiking neural network (SNN) circuit. Through the performance evaluation of the proposed STCLN model for SNN, the optimal ratio of the excitatory and inhibitory synapse is derived.
- 1 January 2024
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 15 (4) , 796-810
Open AccessSpatiotemporal contextual learning network with excitatory and inhibitory synapses for spiking neural network hardware
A spatiotemporal contextual learning network (STCLN) model is suitable for edge applications. To implement the STCLN model as a small analog asynchronous integrated circuit with low power consumption, this study proposes to introduce the excitatory and inhibitory synapses into the STCLN model which can be implemented by an event-driven spiking neural network (SNN) circuit. Through the performance evaluation of the proposed STCLN model for SNN, the optimal ratio of the excitatory and inhibitory synapse is derived.
- Open Access
- 1 January 2024
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 15 (4) , 750-763
Effect of memory capacity characteristics on time-series prediction performance of reservoir neural network with extended chaotic neural network model
We examined the memory capacity (MC) characteristics of a reservoir neural network (RNN) that uses an extended chaotic neural network (ExCNN) model as the reservoir layer. Furthermore, we propose a novel prediction performance measure based on the long tail property of MC. To design RNN using the ExCNN model, we derived the echo state property of the ExCNN model. To evaluate the effect of exponential local memory terms of the ExCNN model on prediction performance, we performed a closed-loop one-step prediction of a periodic random sequence. The results indicate that RNN using the ExCNN model with high prediction performance exhibits a distinctive long tail in MC characteristic curve with respect to a calculation delay.
- 1 January 2024
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 15 (4) , 750-763
Open AccessEffect of memory capacity characteristics on time-series prediction performance of reservoir neural network with extended chaotic neural network model
We examined the memory capacity (MC) characteristics of a reservoir neural network (RNN) that uses an extended chaotic neural network (ExCNN) model as the reservoir layer. Furthermore, we propose a novel prediction performance measure based on the long tail property of MC. To design RNN using the ExCNN model, we derived the echo state property of the ExCNN model. To evaluate the effect of exponential local memory terms of the ExCNN model on prediction performance, we performed a closed-loop one-step prediction of a periodic random sequence. The results indicate that RNN using the ExCNN model with high prediction performance exhibits a distinctive long tail in MC characteristic curve with respect to a calculation delay.
- 1 January 2022
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 13 (2) , 306-311
Preliminary experimental results of a stacked 3D cyclic chaotic neural network reservoir integrated circuit
A cyclic chaotic neural network reservoir (CNNR) circuit is implemented on a stacked three-dimensional (3D) cyclic neural network integrated circuit (IC) system, which was originally designed and fabricated for a deep feedforward neural network. We create a circuit emulator for a neuron chip used in the 3D integrated system according to the measurement results obtained from the chip. We preliminary evaluate the performance of the cyclic CNNR circuit for a waveform classification task using the emulator.
- 1 January 2022
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 13 (2) , 306-311
Preliminary experimental results of a stacked 3D cyclic chaotic neural network reservoir integrated circuit
A cyclic chaotic neural network reservoir (CNNR) circuit is implemented on a stacked three-dimensional (3D) cyclic neural network integrated circuit (IC) system, which was originally designed and fabricated for a deep feedforward neural network. We create a circuit emulator for a neuron chip used in the 3D integrated system according to the measurement results obtained from the chip. We preliminary evaluate the performance of the cyclic CNNR circuit for a waveform classification task using the emulator.
- 1 January 2022
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 13 (2) , 197-202
Designing the human-centric IoT society: Cooperative industry-academic strategies for creative future connection
This paper summarizes the open panel discussions at the special session (SS) held on December 8th, 2021, entitled “Designing the human-centric IoT society: Cooperative industry-academic strategies for creative future connection” in the 2021 Nonlinear Science Workshop (NLSW2021). The SS was jointly organized by Tohoku Forum for Creativity(TFC) and the NOLTA society. Before the SS, an online advance questionnaire survey was conducted using Google Forms. After three introductory talks, 52 participants from academic and industry fields discussed, and shared their idea in the open panel for the future IoT society to enhance human well-being.
- 1 January 2022
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in Nonlinear Theory and Its Applications, IEICE
- Vol. 13 (2) , 197-202
Designing the human-centric IoT society: Cooperative industry-academic strategies for creative future connection
This paper summarizes the open panel discussions at the special session (SS) held on December 8th, 2021, entitled “Designing the human-centric IoT society: Cooperative industry-academic strategies for creative future connection” in the 2021 Nonlinear Science Workshop (NLSW2021). The SS was jointly organized by Tohoku Forum for Creativity(TFC) and the NOLTA society. Before the SS, an online advance questionnaire survey was conducted using Google Forms. After three introductory talks, 52 participants from academic and industry fields discussed, and shared their idea in the open panel for the future IoT society to enhance human well-being.
- 1 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Design for 3-D Stacked Neural Network Circuit with Cyclic Analog Computing
An analog-based CMOS neuron with crossbar configuration multiplier-adder circuit for three-dimensional chip stacking (3-D) cyclic signal processing deep neural network is presented. The circuit designed to verify the proposed concept is a total of four-layer stacked chips, consisting of two neuron chips that mainly perform analog multiply-accumulate (MAC) operations and two memory chips that store neuron potential (activation) and weight data. According to papers and conference presentations on accelerators for the neural network circuit, the analog MAC operation has the potential to achieve high efficiency by using the analog processing of the crossbar configuration. However, analog MAC circuit with compensation technique is required to overcome the decrease in computing accuracy due to electrical noise and element nonlinearity. In this study, we have designed and verified a highly linearity multiplier-adder circuit for highly efficient 3-D cyclic analog processing. In this paper, we focus on the current adder circuit with switched capacitor correlated double sampling (CDS) technique, to minimize non-linearity, noise, and circuit area. The proposed the analog adder circuit was designed with 180-nm and operation power supply voltage 1.8V CMOS technology, and the fabricated neuron chip that integrates 64 neurons and $64 \times 64$ synaptic connections. As analysis results, it was confirmed that the CDS technique apply analog adder circuit reduced the noise output voltage less than 0.1mV and the range of voltage linearity was over 1.6V.- 1 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Design for 3-D Stacked Neural Network Circuit with Cyclic Analog Computing
An analog-based CMOS neuron with crossbar configuration multiplier-adder circuit for three-dimensional chip stacking (3-D) cyclic signal processing deep neural network is presented. The circuit designed to verify the proposed concept is a total of four-layer stacked chips, consisting of two neuron chips that mainly perform analog multiply-accumulate (MAC) operations and two memory chips that store neuron potential (activation) and weight data. According to papers and conference presentations on accelerators for the neural network circuit, the analog MAC operation has the potential to achieve high efficiency by using the analog processing of the crossbar configuration. However, analog MAC circuit with compensation technique is required to overcome the decrease in computing accuracy due to electrical noise and element nonlinearity. In this study, we have designed and verified a highly linearity multiplier-adder circuit for highly efficient 3-D cyclic analog processing. In this paper, we focus on the current adder circuit with switched capacitor correlated double sampling (CDS) technique, to minimize non-linearity, noise, and circuit area. The proposed the analog adder circuit was designed with 180-nm and operation power supply voltage 1.8V CMOS technology, and the fabricated neuron chip that integrates 64 neurons and $64 \times 64$ synaptic connections. As analysis results, it was confirmed that the CDS technique apply analog adder circuit reduced the noise output voltage less than 0.1mV and the range of voltage linearity was over 1.6V.- 1 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Implementation of a Chaotic Neural Network Reservoir on a TSV/$\mu\text{Bump}$ Stacked 3D Cyclic Neural Network Integrated Circuit
A TSV/ $\mu\text{Bump}$ stacked 3D cyclic deep neural network integrated circuit architecture was proposed. Furthermore, a technique for embedding a chaotic neural network reservoir into the proposed architecture was devised. A proof-of-concept neural network chip, and a weight memory chip have been designed and fabricated to confirm the feasibly of the proposed architecture. In this study, the neuron chip configured as a cyclic chaotic neuron circuit is evaluated by measuring circuit building blocks, and constructing a dedicated MATLAB circuit emulator based on the measurement results from the chip. Some of the results from the MATLAB emulator are illustrated.- 1 October 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Implementation of a Chaotic Neural Network Reservoir on a TSV/$\mu\text{Bump}$ Stacked 3D Cyclic Neural Network Integrated Circuit
A TSV/ $\mu\text{Bump}$ stacked 3D cyclic deep neural network integrated circuit architecture was proposed. Furthermore, a technique for embedding a chaotic neural network reservoir into the proposed architecture was devised. A proof-of-concept neural network chip, and a weight memory chip have been designed and fabricated to confirm the feasibly of the proposed architecture. In this study, the neuron chip configured as a cyclic chaotic neuron circuit is evaluated by measuring circuit building blocks, and constructing a dedicated MATLAB circuit emulator based on the measurement results from the chip. Some of the results from the MATLAB emulator are illustrated.