Filters








4,624 Hits in 8.8 sec

Event-Based Angular Velocity Regression with Spiking Networks [article]

Mathias Gehrig, Sumit Bam Shrestha, Daniel Mouritzen, Davide Scaramuzza
2020 arXiv   pre-print
Spiking Neural Networks (SNNs) are bio-inspired networks that process information conveyed as temporal spikes rather than numeric values.  ...  We propose, for the first time, a temporal regression problem of numerical values given events from an event camera.  ...  designed for regression of numeric values. • A detailed evaluation against state-of-the-art ANN models crafted for event-based vision problems.  ... 
arXiv:2003.02790v1 fatcat:ako4zieq4vg6fhf5cuq7w3psti

Deep Spiking Convolutional Neural Network for Single Object Localization Based On Deep Continuous Local Learning [article]

Sami Barchid, José Mennesson, Chaabane Djéraba
2021 arXiv   pre-print
We propose a deep convolutional spiking neural network for the localization of a single object in a grayscale image.  ...  The encouraging results reported on Oxford-IIIT-Pet validates the exploitation of spiking neural networks with a supervised learning approach for more elaborate vision tasks in the future.  ...  INTRODUCTION Computer vision has shown great progress with the advent of Artificial Neural Networks (ANN) and deep learning, which achieves state-of-the-art performance for most vision tasks [1] .  ... 
arXiv:2105.05609v1 fatcat:7eanah447bfa3o3q5xsqzrufp4

Deep Spiking Convolutional Neural Network for Single Object Localization Based On Deep Continuous Local Learning

Sami Barchid, Jose Mennesson, Chaabane Djeraba
2021 2021 International Conference on Content-Based Multimedia Indexing (CBMI)  
We propose a deep convolutional spiking neural network for the localization of a single object in a grayscale image.  ...  The encouraging results reported on Oxford-IIIT-Pet validates the exploitation of spiking neural networks with a supervised learning approach for more elaborate vision tasks in the future.  ...  INTRODUCTION Computer vision has shown great progress with the advent of Artificial Neural Networks (ANN) and deep learning, which achieves state-of-the-art performance for most vision tasks [1] .  ... 
doi:10.1109/cbmi50038.2021.9461880 fatcat:vagitf4b6vh6datdgv3ppjba4m

Editorial Biologically Learned/Inspired Methods for Sensing, Control, and Decision

Yongduan Song, Jennie Si, Sonya Coleman, Dermot Kerr
2022 IEEE Transactions on Neural Networks and Learning Systems  
Evaluations using the MNIST dataset show an accuracy of 98.5% in the case of the fully connected SNN and 99.4% with the C-SNN, which is the state-of-the-art in spike-driven learning algorithms for DeepSNNs  ...  The experiment results show that MeRec outperforms previous state-of-the-art approaches with at least 50% accuracy drop reduction for several compared tasks.  ... 
doi:10.1109/tnnls.2022.3161003 fatcat:4e6v2kclcbb5pgkqqsyyaiwzjy

Neuromorphic Design Using Reward-based STDP Learning on Event-Based Reconfigurable Cluster Architecture

Mahyar Shahsavari, David Thomas, Andrew Brown, Wayne Luk
2021 International Conference on Neuromorphic Systems 2021  
Neuromorphic computing systems simulate spiking neural networks that are used for research into how biological neural networks function, as well as for applied engineering such as robotics, pattern recognition  ...  We evaluate the system performance in a single box of our designed architecture using 6000 concurrent hardware threads and demonstrate scaling to networks with up to 2 million neurons and 400 million synapses  ...  ACKNOWLEDGMENTS Thanks to Jonathan Beaumont and Matthew Naylor for their technical development supports. This work is sponsored by UK EPSRC grant EP/N031768/1 (POETS project).  ... 
doi:10.1145/3477145.3477151 fatcat:2onvwqn5crdjbopmhi6vwd44uq

SPICEprop: Backpropagating Errors Through Memristive Spiking Neural Networks [article]

Peng Zhou, Jason K. Eshraghian, Dong-Uk Choi, Sung-Mo Kang
2022 arXiv   pre-print
The natural spiking dynamics of the MIF neuron model are fully differentiable, eliminating the need for gradient approximations that are prevalent in the spiking neural network literature.  ...  We present a fully memristive spiking neural network (MSNN) consisting of novel memristive neurons trained using the backpropagation through time (BPTT) learning rule.  ...  We achieve state of the art accuracy on MNIST and Fashion-MNIST datasets when compared to all other networks of spiking MSNNs.  ... 
arXiv:2203.01426v3 fatcat:bvii3xz73fdopbmoddxwa2d2qu

Learning and Real-time Classification of Hand-written Digits With Spiking Neural Networks [article]

Shruti R. Kulkarni, John M. Alexiades, Bipin Rajendran
2017 arXiv   pre-print
On the standard MNIST database images of handwritten digits, our network achieves an accuracy of 99.80% on the training set and 98.06% on the test set, with nearly 7x fewer parameters compared to the state-of-the-art  ...  based supervised learning algorithm for spiking neural networks to adjust the synaptic weights.  ...  The highest accuracy SNN for the MNIST was reported in [16] , where a two-stage convolution neural network achieved an accuracy of 99.31% on the test set.  ... 
arXiv:1711.03637v1 fatcat:d6hqts2vvfaxnlfdk7jm245awm

EDHA: Event-Driven High Accurate Simulator for Spike Neural Networks

Lingfei Mo, Xinao Chen, Gang Wang
2021 Electronics  
However, most of the existing simulators for spike neural networks are clock-driven, which has two main problems.  ...  In order to solve these problems, an event-driven high accurate simulator named EDHA (Event-Driven High Accuracy) for spike neural networks is proposed in this paper.  ...  Acknowledgments: The authors would like to thank all the members of FutureX LAB of Southeast University for their help and support, especially for the robust discussion.  ... 
doi:10.3390/electronics10182281 fatcat:2nns64cfhre2xkcbplnmkwxt5a

Exploring Optimized Spiking Neural Network Architectures for Classification Tasks on Embedded Platforms

Tehreem Syed, Vijay Kakani, Xuenan Cui, Hakil Kim
2021 Sensors  
In this current study, the training is carried out using deep convolutional spiking neural networks with surrogate gradient descent backpropagation in a customized layer architecture similar to deep artificial  ...  neural networks.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/s21093240 pmid:34067080 fatcat:hsknxhxkavaqhg54lylp6hg7wy

Benchmarking Spike-Based Visual Recognition: A Dataset and Evaluation

Qian Liu, Garibaldi Pineda-García, Evangelos Stromatias, Teresa Serrano-Gotarredona, Steve B. Furber
2016 Frontiers in Neuroscience  
ACKNOWLEDGMENTS The work presented in this paper was largely inspired by discussions at the 2015 Workshops on Neuromorphic Cognition Engineering in CapoCaccia.  ...  The authors would also like to thank Patrick Camilleri, Michael Hopkins, and Viv Woods for meaningful discussions and proof-reading the paper.  ...  of the state of the art in spike-based visual recognition, and (4) help researchers identify future directions and advance the field.  ... 
doi:10.3389/fnins.2016.00496 pmid:27853419 pmcid:PMC5090001 fatcat:6qmd7ealdjhjtkcdo6w7lrpury

WaveSense: Efficient Temporal Convolutions with Spiking Neural Networks for Keyword Spotting [article]

Philipp Weidel, Sadique Sheik
2021 arXiv   pre-print
The results show that the proposed network beats the state of the art of other spiking neural networks and reaches near state-of-the-art performance of artificial neural networks such as CNNs and LSTMs  ...  Neuromorphic processors emulating spiking neural networks show great computational power while fulfilling the limited power budget as needed in this domain.  ...  Sec. 2 details all the methods used for audio data pre-processing, conversion to spikes and the details of the network architecture.  ... 
arXiv:2111.01456v1 fatcat:akvq2s6szngrvja6rhmaihayzq

Spiking Neural Networks Hardware Implementations and Challenges

Maxence Bouvier, Alexandre Valentian, Thomas Mesquida, Francois Rummens, Marina Reyboz, Elisa Vianello, Edith Beigne
2019 ACM Journal on Emerging Technologies in Computing Systems  
In this survey, we present the state of the art of hardware implementations of spiking neural networks and the current trends in algorithm elaboration from model selection to training mechanisms.  ...  They are expected to improve the computational performance and efficiency of neural networks, but are best suited for hardware able to support their temporal dynamics.  ...  Through simulation of a spiking convolutional neural network using the same computational units, they reach 99.40% accuracy, which puts them slightly behind the state of the art [81, 155] .  ... 
doi:10.1145/3304103 fatcat:p3frra3osnhybj4hkor4y5cyqm

Reducing the computational footprint for real-time BCPNN learning

Bernhard Vogginger, René Schüffny, Anders Lansner, Love Cederström, Johannes Partzsch, Sebastian Höppner
2015 Frontiers in Neuroscience  
Aiming for a small memory footprint per BCPNN synapse, we also evaluate the use of fixed-point numbers for the state variables, and assess the number of bits required to achieve same or better accuracy  ...  Ultimately, in a typical use case, the simulation using our approach is more than one order of magnitude faster than with the fixed step size Euler method.  ...  ACKNOWLEDGMENTS We would like to thank Édi Kettemann for his work on the analytical I solution.  ... 
doi:10.3389/fnins.2015.00002 pmid:25657618 pmcid:PMC4302947 fatcat:33jrywk2mnhkjlonxpelmszuqu

Exploring the Effects of Caputo Fractional Derivative in Spiking Neural Network Training

Natabara Máté Gyöngyössy, Gábor Eros, János Botzheim
2022 Electronics  
Fractional calculus is an emerging topic in artificial neural network training, especially when using gradient-based methods.  ...  This paper brings the idea of fractional derivatives to spiking neural network training using Caputo derivative-based gradient calculation.  ...  To reach state-of-the-art performance on MNIST more complex deep neural networks have to be used with refined temporal simulation.  ... 
doi:10.3390/electronics11142114 fatcat:lv5hvponrratze3iyg2rohlt6i

A Predictive Model for Student Achievement Using Spiking Neural Networks Based on Educational Data

Chuang Liu, Haojie Wang, Yingkui Du, Zhonghu Yuan
2022 Applied Sciences  
The experimental results show that the model based on spiking neural network can effectively improve the prediction accuracy of student achievement.  ...  to evaluate the quality of courses.  ...  With the further development of future research, research in related fields will inevitably make greater breakthroughs, and achieve greater results for the application of data mining in the field of education  ... 
doi:10.3390/app12083841 fatcat:nh7ggjbztnczbl4mdz2kzwkxdq
« Previous Showing results 1 — 15 out of 4,624 results