Filters








275 Hits in 2.8 sec

Brain-Inspired Hardware for Artificial Intelligence: Accelerated Learning in a Physical-Model Spiking Neural Network [chapter]

Timo Wunderlich, Akos F. Kungl, Eric Müller, Johannes Schemmel, Mihai Petrovici
2019 Lecture Notes in Computer Science  
We present results from a prototype chip of the BrainScaleS-2 mixed-signal neuromorphic system that adopts a physical-model approach with a 1000-fold acceleration of spiking neural network dynamics relative  ...  The experiment demonstrates key aspects of the employed approach, such as accelerated and flexible learning, high energy efficiency and resilience to noise.  ...  We present results from a prototype chip of the BrainScaleS-2 mixed-signal neuromorphic system that adopts a physical-model approach with a 1000-fold acceleration of spiking neural network dynamics relative  ... 
doi:10.1007/978-3-030-30487-4_10 fatcat:hyms63nkwrc3rm7mvttrtzmx34

Editorial: Robust Artificial Intelligence for Neurorobotics

Joe Hays, Subramanian Ramamoorthy, Christian Tetzlaff
2021 Frontiers in Neurorobotics  
SR acknowledges support from the US Office of Naval Research-Global (award no. N62909-19-1-2072) for  ...  Think Three articles focused on general approaches to algorithm development based on Spiking Neural Networks, representing the thinking aspects of systems.  ...  OUTLOOK FOR THE FUTURE A long standing open question at the intersection of many fields-Artificial Intelligence, Neural Computing, Neuromorphic Systems and other forms of Biomimesis-pertains to the specification  ... 
doi:10.3389/fnbot.2021.809903 pmid:34975448 pmcid:PMC8716384 fatcat:za4rkweh5fdbzlg3r6pco7kft4

Neural Coding in Spiking Neural Networks: A Comparative Study for Robust Neuromorphic Systems

Wenzhe Guo, Mohammed E Fouda, Ahmed M Eltawil, Khaled Nabil Salama
2021 Frontiers in Neuroscience  
Furthermore, the robustness of each coding scheme to the non-ideality-induced synaptic noise and fault in analog neuromorphic systems was studied and compared.  ...  and considerations in neuromorphic systems.  ...  For example, noise resilience, fault tolerance, and implementation overhead are essential considerations in designing a real-time neuromorphic system.  ... 
doi:10.3389/fnins.2021.638474 pmid:33746705 pmcid:PMC7970006 fatcat:s2dli4uiwzerdb6dvbwrwiipke

Emerging Hardware Techniques and EDA Methodologies for Neuromorphic Computing (Dagstuhl Seminar 19152)

Krishnendu Chakrabarty, Tsung-Yi Ho, Hai Li, Ulf Schlichtmann, Michael Wagner
2019 Dagstuhl Reports  
This report documents the program and the outcomes of Dagstuhl Seminar 19152 "Emerging Hardware Techniques and EDA Methodologies for Neuromorphic Computing," which was held during April 7-10, 2019 in Schloss  ...  Seminar April 7-10, 2019 -http://www.dagstuhl.de/19152 2012 ACM Subject Classification Computer systems organization → Neural networks, Hardware → Biology-related information processing, Hardware → Hardware-software  ...  It has the potential to achieve low cost, high noise resilience, and high energy efficiency due to the distributed nature of neural computation and the use of low energy spikes for information exchange  ... 
doi:10.4230/dagrep.9.4.43 dblp:journals/dagstuhl-reports/ChakrabartyH0S19 fatcat:7fpavhm4gzgxnj2o23jm66sjiy

Evolving Spiking Circuit Motifs Using Weight Agnostic Neural Networks

Abrar Anwar
2021 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Neural architecture search (NAS) has emerged as an algorithmic method of developing neural network architectures. Weight Agnostic Neural Networks (WANNs) are an evolutionary-based NAS approach.  ...  Here, we extend the WANN framework to search for spiking circuits and in doing so investigate whether these circuit motifs can also yield task performance that is weight agnostic.  ...  ., for the U.S.  ... 
doi:10.1609/aaai.v35i18.17974 fatcat:xigpcmghprafpf4jxsuoayyiga

Robustness from structure: Inference with hierarchical spiking networks on analog neuromorphic hardware

Mihai A. Petrovici, Anna Schroeder, Oliver Breitwieser, Andreas Grubl, Johannes Schemmel, Karlheinz Meier
2017 2017 International Joint Conference on Neural Networks (IJCNN)  
We then argue that hierarchical networks of leaky integrate-and-fire neurons can offer the required robustness for physical implementation and demonstrate this with both software simulations and emulation  ...  How spiking networks are able to perform probabilistic inference is an intriguing question, not only for understanding information processing in the brain, but also for transferring these computational  ...  MAP and AS designed and performed the experiments in software and on the Spikey chip. OB designed the software environment for performing the simulations.  ... 
doi:10.1109/ijcnn.2017.7966123 dblp:conf/ijcnn/PetroviciSBGSM17 fatcat:hexp66obkjcmfmbhqjjzs5kmda

A cross-layer approach to cognitive computing

Gobinda Saha, Cheng Wang, Anand Raghunathan, Kaushik Roy
2022 Proceedings of the 59th ACM/IEEE Design Automation Conference  
This effort spans new learning algorithms inspired from biological information processing principles, network architectures best suited for such algorithms, and neuromorphic hardware substrates such as  ...  Therefore, it is imperative to search for fundamentally new approaches so that the improvement in computing performance and efficiency can keep up with the exponential growth of the AI computational demand  ...  for robust machine vision systems.  ... 
doi:10.1145/3489517.3530642 fatcat:iflcowivyvchriny7qcqukua7q

Neuromorphic Electronic Systems for Reservoir Computing [article]

Fatemeh Hadaeghi
2020 arXiv   pre-print
This chapter provides a comprehensive survey of the researches and motivations for hardware implementation of reservoir computing (RC) on neuromorphic electronic systems.  ...  Due to its computational efficiency and the fact that training amounts to a simple linear regression, both spiking and non-spiking implementations of reservoir computing on neuromorphic hardware have been  ...  I would also like to thank Herbert Jaeger, who provided insight and expertise that greatly assisted this research.  ... 
arXiv:1908.09572v2 fatcat:cimkbnvyrjc3lhixlyufgmqy3i

Surrogate gradients for analog neuromorphic computing

Benjamin Cramer, Sebastian Billaudelle, Simeon Kanya, Aron Leibfried, Andreas Grübl, Vitali Karasenko, Christian Pehle, Korbinian Schreiber, Yannik Stradmann, Johannes Weis, Johannes Schemmel, Friedemann Zenke
2022 Proceedings of the National Academy of Sciences of the United States of America  
Surrogate gradient learning has emerged as a promising training strategy for spiking networks, but its applicability for analog neuromorphic systems has not been demonstrated.  ...  Analog neuromorphic hardware uses the same principles to emulate spiking neural networks with exceptional energy efficiency.  ...  Baumbach for his valuable feedback during the early commissioning phase of the system; and J. Göltz and L. Kriener for helpful discussions.  ... 
doi:10.1073/pnas.2109194119 pmid:35042792 pmcid:PMC8794842 fatcat:j3x7ocfndjanrhxys6rbicqrgi

Surrogate gradients for analog neuromorphic computing [article]

Benjamin Cramer, Sebastian Billaudelle, Simeon Kanya, Aron Leibfried, Andreas Grübl, Vitali Karasenko, Christian Pehle, Korbinian Schreiber, Yannik Stradmann, Johannes Weis, Johannes Schemmel, Friedemann Zenke
2021 arXiv   pre-print
Using the BrainScaleS-2 neuromorphic system, we show that learning self-corrects for device mismatch resulting in competitive spiking network performance on both vision and speech benchmarks.  ...  Analog neuromorphic hardware uses the same principles to emulate spiking neural networks with exceptional energy-efficiency.  ...  Baumbach for his valuable feedback during the early commissioning phase of the system, and J. Göltz and L. Kriener for helpful discussions.  ... 
arXiv:2006.07239v3 fatcat:5n4kkttobrfjfkmydjgtrmwila

Autonomous Flying With Neuromorphic Sensing

Patricia P. Parlevliet, Andrey Kanaev, Chou P. Hung, Andreas Schweiger, Frederick D. Gregory, Ryad Benosman, Guido C. H. E. de Croon, Yoram Gutfreund, Chung-Chuan Lo, Cynthia F. Moss
2021 Frontiers in Neuroscience  
The implementation of neuromorphic paradigms for autonomous flight will require fundamental changes in both traditional hardware and software.  ...  This results in both energy and computational resource savings being an inspiration for autonomous systems.  ...  Another neuromorphic approach to active sensing, includes the world's first spiking neural network-based chip that was announced recently for radar signal processing.  ... 
doi:10.3389/fnins.2021.672161 pmid:34054420 pmcid:PMC8160287 fatcat:nw6twkp3abeonlkqmtvc3nutsi

Neuromorphic Electronic Circuits for Building Autonomous Cognitive Systems

Elisabetta Chicca, Fabio Stefanini, Chiara Bartolozzi, Giacomo Indiveri
2014 Proceedings of the IEEE  
Several analog and digital brain-inspired electronic systems have been recently proposed as dedicated solutions for fast simulations of spiking neural networks.  ...  properties of recurrent neural networks and show how neuromorphic Winner-Take-All circuits can implement working-memory and decision-making mechanisms.  ...  ; and 2) that synaptic states are resilient to changes due to spontaneous activity, thus increasing the robustness to noise.  ... 
doi:10.1109/jproc.2014.2313954 fatcat:qbbuxacqtvci5ceffnqsjdif3y

Self-Testing Analog Spiking Neuron Circuit

Sarah A. El-Sayed, Luis A. Camunas-Mesa, Bernabe Linares-Barranco, Haralampos-G. Stratigopoulos
2019 2019 16th International Conference on Synthesis, Modeling, Analysis and Simulation Methods and Applications to Circuit Design (SMACD)  
In this paper, we address the problem of post-manufacturing test and self-test of hardware-implemented neural networks. In particular, we propose a self-testable version of a spiking neuron circuit.  ...  Hardware-implemented neural networks are foreseen to play an increasing role in numerous applications.  ...  ACKNOWLEDGMENTS This work has been carried out in the framework of the Penta HADES project. Luis A. Camuñas-Mesa was funded by the VI PPIT through the Universidad de Sevilla.  ... 
doi:10.1109/smacd.2019.8795234 dblp:conf/smacd/El-SayedCLS19 fatcat:txg3ajwwqvevhiage74fpk67l4

Robust neuromorphic coupled oscillators for adaptive pacemakers [article]

Renate Krause
2021 arXiv   pre-print
In this work, we present a robust, spiking neural network model of neural coupled oscillators and validate it with an implementation on a mixed-signal neuromorphic processor.  ...  The advent of mixed-signal analog/digital neuromorphic electronic circuits provides new means for implementing neural coupled oscillators on compact low-power spiking neural network hardware platforms.  ...  The code to tune and run the system of coupled oscillators on the DYNAP-SE board is available at: https://gitlab.com/rekrau/neuromorphicOscillatorsForPacemakers. Acknowledgements  ... 
arXiv:2104.01638v1 fatcat:nkigu24ovjeixm7ld5uo5rxewu

Memristors Empower Spiking Neurons With Stochasticity

Maruan Al-Shedivat, Rawan Naous, Gert Cauwenberghs, Khaled Nabil Salama
2015 IEEE Journal on Emerging and Selected Topics in Circuits and Systems  
neural networks.  ...  The simulations and analysis of the behavior of a single stochastic neuron and a winner-take-all network built of such neurons and trained on handwritten digits confirm that the circuit can be used for  ...  Neftci and M. Affan Zidan for helpful discussions and advice.  ... 
doi:10.1109/jetcas.2015.2435512 fatcat:yfloem2wx5gktol7edhpgxp7ru
« Previous Showing results 1 — 15 out of 275 results