A data-based large-scale model for primary visual cortex enables brain-like robust and versatile visual processing

Chen, Guozhang; Scherr, Franz; Maass, Wolfgang

doi:10.1126/sciadv.abq7592

“…SpikingJelly 5 incorporates many neuromorphic datasets like CIFAR10-DVS [112], ASL-DVS [113], and DVS Gesture [114], which can be easily loaded by users. Norse [109] tries to introduce the sparse and event-driven characteristics of SNNs and supports many typical neuron [87] University of Heidelberg 2010 Python Neurogrid [93] Stanford University 2014 Python Darwin [83] Zhejiang University 2015 Customized Language Darwin-MDL TrueNorth [40] IBM 2015 Corelet Loihi [41] Intel 2018 Python Braindrop [104] Stanford University 2018 Python SpiNNaker [105] University of Manchester 2018 Python Tianjic [85], [86] Tsinghua University 2019 C++, Matlab Loihi 2 [106] Intel 2021 Python BrainScaleS2 [82] Heidelberg Peking University, Peng Cheng Laboratory 2020 Python Norse [117] University of Heidelberg 2021 Python SNNtorch [110] University of Michigan 2021 Python Brain Network Simulators GENESIS [118] California Institute of Technology 1988 Customized language Neuron [119] Yale University 2006 Customized Language, Python NEST [120] The NEST Initiative 2007 Customized Language SLI, Python, C++ Brian [121] Ecole Normale Superieure 2008 Python PyNN [122] CNRS 2009 Python TVB [123] Aix Marseille Université 2013 Python, Matlab Brian2 [124] Ecole Normale Supérieure 2014 Python Auryn [125] Ecole Polytechnique Fédérale de Lausanne 2014 C++ ANNarchy [126] Technische Universität Chemnitz 2015 C++ GeNN [127] University of Sussex 2016 Python, C++ Brian2GeNN [128] University of Sussex 2020 Python,C++ BSIM [129] Tsinghua University 2020 Python, C++ MONET [130] RIKEN 2020 Python, C BrainPy [131] Peking University 2022 Python V1 Simulator [133] Graz University of Technology 2022 Python (Tensorflow) BrainCog…”

Section: B Algorithm Programming Platformsmentioning

confidence: 99%

“…On the software tool side, current brain-inspired software can be partitioned into three categories according to their usage and infrastructure: neuromorphic toolchains [40], [41], [82], [83], [85]- [87], [93], [104]- [106], algorithm programming platforms [107]- [117] and brain network simulators [118]- [133]. Neuromorphic toolchains are often concurrently designed for the specific chips, which aim to facilitate the high-level model designing and compile programs to the lowlevel executable codes described by computation primitives supported by the chips.…”

Section: Introduction Motivation and Overviewmentioning

confidence: 99%

Brain Inspired Computing: A Systematic Survey and Future Trends

Li¹,

Deng²,

Tang³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Brain Inspired Computing (BIC) is an emerging research field that aims to build fundamental theories, models, hardware architectures, and application systems toward more general Artificial Intelligence (AI) by learning from the information processing mechanisms or structures/functions of biological nervous systems. It is regarded as one of the most promising research directions for future intelligent computing in the post-Moore era. In the past few years, various new schemes in this field have sprung up to explore more general AI. These works are quite divergent in the aspects of modeling/algorithm, software tool, hardware platform, and benchmark data, since BIC is an interdisciplinary field that consists of many different domains, including computational neuroscience, artificial intelligence, computer science, statistical physics, material science, microelectronics and so forth. This situation greatly impedes researchers from obtaining a clear picture and getting started in the right way. Hence, there is an urgent requirement to do a comprehensive survey in this field to help correctly recognize and analyze such bewildering methodologies. What are the key issues to enhance the development of BIC? What roles do the current mainstream technologies play in the general framework of BIC? Which techniques are truly useful in real-world applications? These questions largely remain open. To address the above issues, in this survey we first clarify the biggest challenge of BIC: how can AI models benefit from the recent advancements in computational neuroscience? With this challenge in mind, we will focus on discussing the concept of BIC and summarize four components of BIC infrastructure development: 1) modeling/algorithm; 2) hardware platform; 3) software tool; and 4) benchmark data. For each component, we will summarize its recent progress, main challenges to resolve, and future trends. On the basis of these studies, we present a general framework for the real-world applications of BIC systems, which is promising to benefit both AI and brain science. Finally, we claim that it is extremely important to build a research ecology to promote prosperity continuously in this field.</p>

show abstract

“…SpikingJelly 5 incorporates many neuromorphic datasets like CIFAR10-DVS [112], ASL-DVS [113], and DVS Gesture [114], which can be easily loaded by users. Norse [109] tries to introduce the sparse and event-driven characteristics of SNNs and supports many typical neuron [87] University of Heidelberg 2010 Python Neurogrid [93] Stanford University 2014 Python Darwin [83] Zhejiang University 2015 Customized Language Darwin-MDL TrueNorth [40] IBM 2015 Corelet Loihi [41] Intel 2018 Python Braindrop [104] Stanford University 2018 Python SpiNNaker [105] University of Manchester 2018 Python Tianjic [85], [86] Tsinghua University 2019 C++, Matlab Loihi 2 [106] Intel 2021 Python BrainScaleS2 [82] Heidelberg Peking University, Peng Cheng Laboratory 2020 Python Norse [117] University of Heidelberg 2021 Python SNNtorch [110] University of Michigan 2021 Python Brain Network Simulators GENESIS [118] California Institute of Technology 1988 Customized language Neuron [119] Yale University 2006 Customized Language, Python NEST [120] The NEST Initiative 2007 Customized Language SLI, Python, C++ Brian [121] Ecole Normale Superieure 2008 Python PyNN [122] CNRS 2009 Python TVB [123] Aix Marseille Université 2013 Python, Matlab Brian2 [124] Ecole Normale Supérieure 2014 Python Auryn [125] Ecole Polytechnique Fédérale de Lausanne 2014 C++ ANNarchy [126] Technische Universität Chemnitz 2015 C++ GeNN [127] University of Sussex 2016 Python, C++ Brian2GeNN [128] University of Sussex 2020 Python,C++ BSIM [129] Tsinghua University 2020 Python, C++ MONET [130] RIKEN 2020 Python, C BrainPy [131] Peking University 2022 Python V1 Simulator [133] Graz University of Technology 2022 Python (Tensorflow) BrainCog…”

Section: B Algorithm Programming Platformsmentioning

confidence: 99%

“…It provides a general-purpose programming framework to ease and facilitate user development based on the just-in-time (JIT) compilation. The work in [133] utilizes the common deep learning framework TensorFlow to simulate biologically detailed large-scale models for area V1, termed as "V1 Simulator", on a single GPU.…”

Section: Brain Network Simulatorsmentioning

confidence: 99%

“…On the software tool side, current brain-inspired software can be partitioned into three categories according to their usage and infrastructure: neuromorphic toolchains [40], [41], [82], [83], [85]- [87], [93], [104]- [106], algorithm programming platforms [107]- [117] and brain network simulators [118]- [133]. Neuromorphic toolchains are often concurrently designed for the specific chips, which aim to facilitate the high-level model designing and compile programs to the lowlevel executable codes described by computation primitives supported by the chips.…”

Section: Introduction Motivation and Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Brain Inspired Computing: A Systematic Survey and Future Trends

Li¹,

Deng²,

Tang³

et al. 2023

Preprint

1

0

View full text Add to dashboard Cite

<p>Brain Inspired Computing (BIC) is an emerging research field that aims to build fundamental theories, models, hardware architectures, and application systems toward more general Artificial Intelligence (AI) by learning from the information processing mechanisms or structures/functions of biological nervous systems. It is regarded as one of the most promising research directions for future intelligent computing in the post-Moore era. In the past few years, various new schemes in this field have sprung up to explore more general AI. These works are quite divergent in the aspects of modeling/algorithm, software tool, hardware platform, and benchmark data, since BIC is an interdisciplinary field that consists of many different domains, including computational neuroscience, artificial intelligence, computer science, statistical physics, material science, microelectronics and so forth. This situation greatly impedes researchers from obtaining a clear picture and getting started in the right way. Hence, there is an urgent requirement to do a comprehensive survey in this field to help correctly recognize and analyze such bewildering methodologies. What are the key issues to enhance the development of BIC? What roles do the current mainstream technologies play in the general framework of BIC? Which techniques are truly useful in real-world applications? These questions largely remain open. To address the above issues, in this survey we first clarify the biggest challenge of BIC: how can AI models benefit from the recent advancements in computational neuroscience? With this challenge in mind, we will focus on discussing the concept of BIC and summarize four components of BIC infrastructure development: 1) modeling/algorithm; 2) hardware platform; 3) software tool; and 4) benchmark data. For each component, we will summarize its recent progress, main challenges to resolve, and future trends. On the basis of these studies, we present a general framework for the real-world applications of BIC systems, which is promising to benefit both AI and brain science. Finally, we claim that it is extremely important to build a research ecology to promote prosperity continuously in this field.</p>

show abstract