Filters








2,989 Hits in 4.5 sec

Understanding the Impact of Neural Variations and Random Connections on Inference

Yuan Zeng, Zubayer Ibne Ferdous, Weixiang Zhang, Mufan Xu, Anlan Yu, Drew Patel, Valentin Post, Xiaochen Guo, Yevgeny Berdichevsky, Zhiyuan Yan
2021 Frontiers in Computational Neuroscience  
Neural variations and dynamics are verified by fitting model parameters with biological experimental results.  ...  This work investigates the impact of neural variations and random connections on inference with learning algorithms.  ...  performed with both the computational and the biophysical models.  ... 
doi:10.3389/fncom.2021.612937 pmid:34163343 pmcid:PMC8215547 fatcat:bggaqfaigbbynfecoutb554s64

Inference with Hybrid Bio-hardware Neural Networks [article]

Yuan Zeng and Zubayer Ibne Ferdous and Weixiang Zhang and Mufan Xu and Anlan Yu and Drew Patel and Xiaochen Guo and Yevgeny Berdichevsky and Zhiyuan Yan
2019 arXiv   pre-print
A novel two-layer bio-hardware hybrid neural network is proposed. The biological layer faithfully models variations of synapses, neurons, and network sparsity in in vitro living neural networks.  ...  On the other hand, simplified multi-layer models of neural networks have shown great success on computational tasks such as image classification and speech recognition.  ...  Conclusion In this paper, a hybrid bio-hardware neural network is proposed and studied using both biophysical and computational models.  ... 
arXiv:1905.11594v2 fatcat:btdcr2mmejb6lgcujm6ssmgpmi

CPU-GPU hybrid platform for efficient spiking neural-network simulation

Francisco Naveros, Niceto R Luque, Jesús A Garrido, Richard R Carrillo, Eduardo Ros
2013 BMC Neuroscience  
[1] and GENESIS [2]), or to simulate neuron models with low degree of biophysical detail within large-scale neural networks (Brian [3] and NEST [4] ).  ...  We have extended EDLUT with the ability of dealing with complex neural models and also with detailed network characteristics such as spike propagation delays.  ...  [1] and GENESIS [2] ), or to simulate neuron models with low degree of biophysical detail within large-scale neural networks (Brian [3] and NEST [4] ).  ... 
doi:10.1186/1471-2202-14-s1-p328 pmcid:PMC3704682 fatcat:enjalwfohzdr5fucki67f2c2pa

Markov Chain Abstractions of Electrochemical Reaction-Diffusion in Synaptic Transmission for Neuromorphic Computing

Margot Wagner, Thomas M. Bartol, Terrence J. Sejnowski, Gert Cauwenberghs
2021 Frontiers in Neuroscience  
hardware for efficient emulation at a very large scale and offers near-equivalence in input-output dynamics while preserving biologically interpretable tunable parameters.  ...  Our work abstracts an existing highly detailed, biophysically realistic 3D reaction-diffusion model of a chemical synapse to a compact internal state space representation that maps onto parallel neuromorphic  ...  as a means toward efficient neuromorphic hardware without biophysical compromise.  ... 
doi:10.3389/fnins.2021.698635 pmid:34912188 pmcid:PMC8667025 fatcat:2fxtmjsyhjai7ofqrxzpxgpgg4

The case for emulating insect brains using anatomical "wiring diagrams" equipped with biophysical models of neuronal activity [article]

Logan Thrasher Collins
2019 arXiv   pre-print
By implementing IBE, large-scale scientific infrastructure may shift towards developing more WBEs and move in the direction of comprehensively understanding neural computation.  ...  At this time, constructing a simulated human brain lacks feasibility due to limited experimental data and limited computational resources.  ...  Neuromorphic hardware architectures may further increase the computational efficiency of IBE.  ... 
arXiv:1812.09362v2 fatcat:6oqs6ill2bf2di44pwbe6guyvq

Organic materials and devices for brain-inspired computing: From artificial implementation to biophysical realism

Yoeri van de Burgt, Paschalis Gkoupidenis
2020 MRS bulletin  
We highlight efforts to mimic brain functions such as spatiotemporal processing, homeostasis, and functional connectivity and emphasize current challenges for efficient neuromorphic computing applications  ...  artificial neural networks as well as interfacing with physiological environments due to their biocompatible nature.  ...  Hardware-based agents, for instance, are still struggling to reach the energy efficiency of the brain. 24 In addition, biorealistic emulation (or biophysical realism) of the neural processing functions  ... 
doi:10.1557/mrs.2020.194 fatcat:nkcur63j4vaxngemsnadctuxiq

A biophysically accurate floating point somatic neuroprocessor

Yiwei Zhang, Jose Nunez-Yanez, Joe McGeehan, Edward Regan, Stephen Kelly
2009 2009 International Conference on Field Programmable Logic and Applications  
Models targeting hardware are traditionally based on fixed point implementations and low precision algorithms which incur a significant loss of information.  ...  Biophysically accurate neuron models have emerged as a very useful tool for neuroscience research.  ...  NEURON [2] and GENESIS [3] , hardware solutions tend to remove the biophysical accuracy at this level due to their complexity.  ... 
doi:10.1109/fpl.2009.5272558 dblp:conf/fpl/ZhangNMRK09 fatcat:27jbwyk5srelpkptsofvhqpacm

GPGPU accelerated simulation and parameter tuning for neuromorphic applications

Kristofor D. Carlson, Michael Beyeler, Nikil Dutt, Jeffrey L. Krichmar
2014 2014 19th Asia and South Pacific Design Automation Conference (ASP-DAC)  
neuron models CARLsim -Fast and efficient CUDA GPU impl.  ...  Efficient coding and the neural representation of value. Ann N Y Acad Sci 1251, 13-32.  ...  • May provide a metric for tuning networks of simulated spiking neurons • May also given insights into how real brain networks process information and achieve homeostasis  ... 
doi:10.1109/aspdac.2014.6742952 dblp:conf/aspdac/CarlsonBDK14 fatcat:votbfyv3bfbcvnepzs77ndwnyq

Biologically compatible neural networks with reconfigurable hardware

Juan Carlos Moctezuma, Joseph P. McGeehan, Jose Luis Nunez-Yanez
2015 Microprocessors and microsystems  
This paper presents a reconfigurable hardware neuro-simulator specifically designed to emulate biophysically accurate and biologically compatible neural networks.  ...  The problem of interconnecting neurons with individual synapses is tackled with a novel synaptic architecture where all incoming synapses are merged efficiently in one single accumulative process without  ...  A neuroprocessor architecture with better performance and latency/area reduction using hardware design methodologies including: efficient exponential operation and novel event-time-driven for the synaptic  ... 
doi:10.1016/j.micpro.2015.09.003 fatcat:nhjuuqwcmncavg7pwxquef7b34

International Neuroscience Initiatives through the Lens of High-Performance Computing

Kristofer E. Bouchard, James B. Aimone, Miyoung Chun, Thomas Dean, Michael Denker, Markus Diesmann, David D. Donofrio, Loren M. Frank, Narayanan Kasthuri, Christof Koch, Oliver Rubel, Horst D. Simon (+2 others)
2018 Computer  
Linking the biophysical reality of bottom-up models with the well-defined computations of top-down models could reveal the biophysical mechanisms of neural computations.  ...  The need to interpret data in place, combined with efficient discoverability across hardware resources, means metadata should be centrally accessible and machine readable.  ... 
doi:10.1109/mc.2018.2141039 fatcat:sgpazlzce5bo5kdq3hcdkvpfua

Neuromorphic modeling abstractions and simulation of large-scale cortical networks

Jeffrey L. Krichmar, Nikil Dutt, Jayram M. Nageswaran, Micah Richert
2011 2011 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)  
We outline key modeling abstractions for the brain and focus on spiking neural network models. We discuss aspects of neuronal processing and computational issues related to modeling these processes.  ...  Biological neural systems are well known for their robust and power-efficient operation in highly noisy environments.  ...  Moreover, spiking models, with their digital signaling and sparse coding, are energy efficient and amenable to hardware application development.  ... 
doi:10.1109/iccad.2011.6105350 dblp:conf/iccad/KrichmarDNR11 fatcat:qlbfkeg6vzappkafmjoxxui2au

A Neuromorphic Digital Circuit for Neuronal Information Encoding Using Astrocytic Calcium Oscillations

Farnaz Faramarzi, Fatemeh Azad, Mahmood Amiri, Bernabé Linares-Barranco
2019 Frontiers in Neuroscience  
Considering highly nonlinear equations of the astrocyte model, linear approximation and single constant multiplication (SCM) techniques are employed for efficient hardware execution while maintaining the  ...  Consequently, hardware realization of astrocytes is important for developing the next generation of bio-inspired computing systems.  ...  FF, FA, and MA would like to thank Mr. Nima Salimi-Nezhad and Mrs. Maryam Rajabalipanah for their valuable assistance.  ... 
doi:10.3389/fnins.2019.00998 pmid:31649494 pmcid:PMC6794439 fatcat:lbrbukbna5bhdkr7tbeltwemaq

Why build a virtual brain? Large-scale neural simulations as jump start for cognitive computing

Matteo Colombo
2016 Journal of experimental and theoretical artificial intelligence (Print)  
I am also grateful to lasha Abzianidze, luigi Acerbi, one anonymous referee and to the editor of Jetai, Eric Dietrich, for their constructive criticisms and helpful suggestions.  ...  The leaky integrate-and-fire model is one of the simplest models of spiking neurons. Given its lack of biophysical detail, the range of phenomena that this model can address is limited.  ...  and programming languages for highly efficient computing systems.  ... 
doi:10.1080/0952813x.2016.1148076 fatcat:rvycy7bbrfea5m6db5uugprl2e

Neuronal Processing, Reconfigurable Neural Networks and Stochastic Computing

Sergey Edward Lyshevski, Vlad Shmerko, Marina Alexandra Lyshevski, Svetlana Yanushchkevich
2008 2008 8th IEEE Conference on Nanotechnology  
This will allow one to synthesize computing hardware (circuits, processing platforms, etc.) guarantying efficient computing and processing.  ...  This paper proposes and studies the premise of three-dimensional (3D) reconfigurable vector neural networks ( 3DV NNs). We research a neurocomputing paradigm to accomplish efficient computing.  ...  With the uncertainties on the cornerstone biophysics of processing by biomolecules and their assemblies, one cannot explicitly specify biophysical phenomena, effects and mechanisms utilized.  ... 
doi:10.1109/nano.2008.216 fatcat:mbk4he33nfhajp7sbipty5jewa

Neuromorphic Engineering: From Neural Systems to Brain-Like Engineered Systems

Francesco Carlo Morabito, Andreas G. Andreou, Elisabetta Chicca
2013 Neural Networks  
In ''Computing with networks of spiking neurons on a biophysically motivated floating-gate based neuromorphic integrated circuit'', by Stephen Brink, Stephen Nease, and Paul Hasler, the authors investigate  ...  Analogies to the behaviour of real biological neural systems are also noted. The alternatives for implementing the same computations are discussed and compared in terms of computational efficiency.  ... 
doi:10.1016/j.neunet.2013.07.001 pmid:23899498 fatcat:37jafps74zfj7ftnrodzpiib5y
« Previous Showing results 1 — 15 out of 2,989 results