Article
Multidisciplinary Sciences
Verena Brehm, Johannes W. Austefjord, Serban Lepadatu, Alireza Qaiumzadeh
Summary: The proposed neuromorphic computing model based on antiferromagnetic domain walls can mimic the behavior of biological neurons and has faster processing speed and more functionalities compared to previous models based on ferromagnetic systems.
SCIENTIFIC REPORTS
(2023)
Article
Computer Science, Artificial Intelligence
Yun Zhang, Hong Qu, Xiaoling Luo, Yi Chen, Yuchen Wang, Malu Zhang, Zefang Li
Summary: In this paper, a Recursive Least Squares-Based Learning Rule (RLSBLR) for Spiking Neural Networks (SNNs) is proposed to generate desired spatio-temporal spike trains. Experimental results in different settings show that the proposed RLSBLR outperforms competitive algorithms in terms of learning accuracy, efficiency, and robustness against noise. The integration of modified synaptic delay learning further improves the learning performance.
Review
Chemistry, Multidisciplinary
Hefei Liu, Yuan Qin, Hung-Yu Chen, Jiangbin Wu, Jiahui Ma, Zhonghao Du, Nan Wang, Jingyi Zou, Sen Lin, Xu Zhang, Yuhao Zhang, Han Wang
Summary: This paper reviews the progress of artificial neuronal devices based on emerging volatile switching materials, focusing on the demonstrated neuron models implemented in these devices and their utilization for computational and sensing applications. Furthermore, it discusses the inspirations from neuroscience and engineering methods to enhance the neuronal dynamics that are yet to be realized in artificial neuronal devices and networks towards achieving the full functionalities of biological neurons.
ADVANCED MATERIALS
(2023)
Article
Multidisciplinary Sciences
Silvan Huerkey, Nelson Niemeyer, Jan-Hendrik Schleimer, Stefanie Ryglewski, Susanne Schreiber, Carsten Duch
Summary: This study identifies a miniaturized circuit solution for the central-pattern-generating (CPG) neural network underlying insect asynchronous flight. The network consists of motoneurons interconnected by electrical synapses that produce network activity splayed out in time instead of synchronized across neurons. This mechanism translates unpatterned premotor input into stereotyped neuronal firing with fixed sequences of cell activation, ensuring stable wingbeat power and is conserved across multiple species.
Article
Computer Science, Artificial Intelligence
Omar Zahra, David Navarro-Alarcon, Silvia Tolu
Summary: A detailed cellular-level forward cerebellar model is developed in this study to achieve the goal of completely mimicking human cognitive and motor behavior. The model considers the characteristics of cerebellar cells and is adjusted using an optimization method. The effectiveness and biological plausibility of the proposed controller are demonstrated in robotic manipulation tasks.
INTERNATIONAL JOURNAL OF NEURAL SYSTEMS
(2022)
Article
Physics, Fluids & Plasmas
D. Chalkiadakis, J. Hizanidis
Summary: This paper revisits the dynamical mechanisms of Josephson Junction-based neurons, revealing complex behaviors relevant for neurocomputation and the design of superconducting neuromorphic devices.
Review
Chemistry, Analytical
Pawel Pietrzak, Szymon Szczesny, Damian Huderek, Lukasz Przyborowski
Summary: Spiking neural networks (SNNs) are gaining interest due to their closer resemblance to actual neural networks in the brain compared to artificial neural networks (ANNs). SNNs have the potential to be more energy efficient on event-driven neuromorphic hardware, leading to cost reduction. However, widely available hardware is still lacking.
Article
Optics
Krzysztof Tyszka, Magdalena Furman, Rafal Mirek, Mateusz Krol, Andrzej Opala, Bartlomiej Seredynski, Jan Suffczynski, Wojciech Pacuski, Michal Matuszewski, Jacek Szczytko, Barbara Pietka
Summary: This paper introduces a new approach to neuromorphic photonics using microcavities as building blocks of optical spiking neurons, and demonstrates its feasibility through experimental results. The research shows that exciton-polaritons exhibit characteristics similar to the Leaky Integrate-and-Fire (LIF) mechanism, enabling ultrafast processing on sub-ns timescales.
LASER & PHOTONICS REVIEWS
(2023)
Review
Physics, Applied
Md Mazharul Islam, Shamiul Alam, Md Shafayat Hossain, Kaushik Roy, Ahmedullah Aziz
Summary: The revolution in artificial intelligence has led to a demand for large storage and data processing capabilities. However, power consumption and hardware overhead pose major challenges for building next-generation AI hardware. Neuromorphic computing has gained attention for its low power consumption, but it still has a long way to go to match the energy efficiency of the human brain. Recent interest has been focused on implementing cryogenic neuromorphic hardware, which offers high speed and low power consumption. This article comprehensively reviews cryogenic neuromorphic hardware, categorizes them hierarchically, and provides a comparative analysis of their performance metrics.
JOURNAL OF APPLIED PHYSICS
(2023)
Article
Neurosciences
Lvhui Hu, Xin Liao
Summary: This article discusses the challenge of extracting useful clues related to delayed feedback signals in machine learning and proposes a membrane voltage slope guided algorithm to address this issue. The experimental results demonstrate that the algorithm achieves excellent performance and serves as a valuable reference for aggregate-label learning on spiking neural networks.
FRONTIERS IN NEUROSCIENCE
(2022)
Article
Engineering, Multidisciplinary
Kevin Linka, Amelie Schafer, Xuhui Meng, Zongren Zou, George Em Karniadakis, Ellen Kuhl
Summary: Understanding real-world dynamical phenomena is challenging, and machine learning has become the go-to technology for analyzing and making decisions based on these phenomena. However, traditional neural networks often ignore the fundamental laws of physics and fail to make accurate predictions. In this study, the combination of neural networks, physics informed modeling, and Bayesian inference is used to integrate data, physics, and uncertainties, improving the predictive potential of neural network models.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2022)
Article
Engineering, Electrical & Electronic
Peng Zhou, Dong-Uk Choi, Wei D. Lu, Sung-Mo Kang, Jason K. Eshraghian
Summary: We present MEMprop, a gradient-based learning method for training fully memristive spiking neural networks. By harnessing the device dynamics to trigger voltage spikes, MEMprop eliminates the need for surrogate gradient methods. The implementation is fully memristive, without the need for additional circuits to implement spiking dynamics, and achieves competitive accuracy on several benchmarks.
IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS
(2022)
Review
Automation & Control Systems
Ben Walters, Mohan V. Jacob, Amirali Amirsoleimani, Mostafa Rahimi Azghadi
Summary: As data processing volume increases, the limitations of traditional computers and the need for more efficient computing methods become evident. Neuromorphic computing mimics the brain's low-power and high-speed computations, making it crucial in the era of big data and artificial intelligence. One significant development in this field is the memristor, a device that exhibits neuromorphic tendencies.
ADVANCED INTELLIGENT SYSTEMS
(2023)
Review
Chemistry, Multidisciplinary
Joon-Kyu Han, Seong-Yun Yun, Sang-Won Lee, Ji-Man Yu, Yang-Kyu Choi
Summary: A spiking neural network (SNN) inspired by the structure and principles of the human brain can greatly improve the energy efficiency of artificial intelligence computing. The construction of artificial neurons at the device level is important for implementing SNNs, and efforts have been focused on developing artificial sensory neurons to reduce hardware cost and energy consumption. This review article surveys recent progress in artificial neuron devices for neural processing and sensing, and discusses the challenges and prospects for their development.
ADVANCED FUNCTIONAL MATERIALS
(2022)
Article
Multidisciplinary Sciences
Jannik Luboeinski, Luis Claro, Andres Pomi, Eduardo Mizraji
Summary: Mechanisms for ensuring stability in dynamical systems are crucial in our interconnected world. The connectivity-stability dilemma suggests that increased connectivity reduces stability, making it challenging to stabilize complex systems while maintaining high connectivity. We found that stabilizing small-world and scale-free networks is possible by increasing self-coupling strength, and the average self-coupling needed for stability increases slower than network size. This has practical implications for stabilizing diverse complex systems.
SCIENTIFIC REPORTS
(2023)
