Hybrid neuromorphic computing,integrating Artificial Neural Networks(ANNs)and Spiking Neural Networks(SNNs),is a key approach to advancing Artificial General Intelligence(AGI).Current hybrid platforms are limited to L...Hybrid neuromorphic computing,integrating Artificial Neural Networks(ANNs)and Spiking Neural Networks(SNNs),is a key approach to advancing Artificial General Intelligence(AGI).Current hybrid platforms are limited to Leaky Integrate-and-Fire(LIF)based SNNs,missing crucial biological neuron behaviors like bursting and adaptation.We propose a hybrid platform based on the TianjicX chip,enabling heterogeneous integration of multiple SNN models(LIF,Quadratic Integrate-and-Fire(QIF),and Izhikevich)alongside ANNs.Our platform employs a co-design strategy for computing and storage mechanisms,minimizing data movement.Simulations show that the co-design approach reduces energy consumption by 8.11%(48.67 mW)compared to TianjicX.The platform also demonstrates superior computational performance across SNN models.It achieves 95%classification accuracy on the MNIST dataset(3000 images,each being 28 pixel×28 pixel and single presentation),surpassing Open Date Index Name(ODIN)by 10.5%.This is achieved with a two-layer fully-connected Izhikevich network(784×800×10),where each synapse operates at 8-bit precision.The network processes 33900 images per second,using only 35 cores(21.88%of 160 cores)and delivering 896 billion operations per second.Furthermore,on ResNet-50,our platform shows a 3.12%increase in computing speed and 40.85 mW/frame reduction in energy consumption compared to the TianjicX chip.展开更多
The hippocampus is thought to contribute largely to memory processing and spatial navigation.Various research projects have shown evidence regarding these two crucial roles.However,many unknown functions of hippocampu...The hippocampus is thought to contribute largely to memory processing and spatial navigation.Various research projects have shown evidence regarding these two crucial roles.However,many unknown functions of hippocampus remain.A great deal of research on the hippocampus is ongoing,but much of this research deals with single neuron,and little research has been conducted on what happens between neurons in hippocampus as they play these roles.In this paper,we intend to examine what changes hippocampal neurons undergo in response to a stimulus.Using an imbalanced,index of timedependent Gini's coefficient,the firing balance between neurons during the moment in which a stimulus is received is examined.More importantly,the different firing balances are observed in reward situation.The result demonstrates the multineuron in hippocampus fires with balance for a while when the rat has reward.In addition,time-dependent Gini's coefficient is a feature that can verify what is not shown by using existing features.展开更多
文摘Hybrid neuromorphic computing,integrating Artificial Neural Networks(ANNs)and Spiking Neural Networks(SNNs),is a key approach to advancing Artificial General Intelligence(AGI).Current hybrid platforms are limited to Leaky Integrate-and-Fire(LIF)based SNNs,missing crucial biological neuron behaviors like bursting and adaptation.We propose a hybrid platform based on the TianjicX chip,enabling heterogeneous integration of multiple SNN models(LIF,Quadratic Integrate-and-Fire(QIF),and Izhikevich)alongside ANNs.Our platform employs a co-design strategy for computing and storage mechanisms,minimizing data movement.Simulations show that the co-design approach reduces energy consumption by 8.11%(48.67 mW)compared to TianjicX.The platform also demonstrates superior computational performance across SNN models.It achieves 95%classification accuracy on the MNIST dataset(3000 images,each being 28 pixel×28 pixel and single presentation),surpassing Open Date Index Name(ODIN)by 10.5%.This is achieved with a two-layer fully-connected Izhikevich network(784×800×10),where each synapse operates at 8-bit precision.The network processes 33900 images per second,using only 35 cores(21.88%of 160 cores)and delivering 896 billion operations per second.Furthermore,on ResNet-50,our platform shows a 3.12%increase in computing speed and 40.85 mW/frame reduction in energy consumption compared to the TianjicX chip.
基金The MSIP(Ministry of Science,ICT&Future Planning),Korea,under the ITRC(Information Technology Research Center)support program(NIPA-2013-H0301-13-2006)supervised by the NIPA(National IT Industry Promotion Agency)The Brain Research Program through the National Research Foundation of Korea funded by the Ministry of Science,ICT&Future Planning(2011-0019212)
文摘The hippocampus is thought to contribute largely to memory processing and spatial navigation.Various research projects have shown evidence regarding these two crucial roles.However,many unknown functions of hippocampus remain.A great deal of research on the hippocampus is ongoing,but much of this research deals with single neuron,and little research has been conducted on what happens between neurons in hippocampus as they play these roles.In this paper,we intend to examine what changes hippocampal neurons undergo in response to a stimulus.Using an imbalanced,index of timedependent Gini's coefficient,the firing balance between neurons during the moment in which a stimulus is received is examined.More importantly,the different firing balances are observed in reward situation.The result demonstrates the multineuron in hippocampus fires with balance for a while when the rat has reward.In addition,time-dependent Gini's coefficient is a feature that can verify what is not shown by using existing features.
