文章目录
- 题目:Deep learning in spiking neural networks
- 时间:2019
- 期刊:Neural Network(B类)
- 研究机构:路易斯安那大学拉法叶分校
名词解释
- spike-timing-dependent plasticity(STDP)
Neuroscientists have identified many variants of this learning rule that falls under the umbrella term spike-timing-dependent plasticity (STDP).
Its key feature is that the weight (synaptic efficacy) connecting a pre- and post-synaptic neuron is adjusted according to their relative spike times within an interval of roughly tens of milliseconds in length
在Scholarpedia 网站中,STDP的定义为
Spike Timing Dependent Plasticity (STDP) is a temporally asymmetric form of Hebbian learning induced by tight temporal correlations between the spikes of pre- and postsynaptic neurons.
As with other forms of synaptic plasticity, it is widely believed that it underlies learning and information storage in the brain, as well as the development and refinement of neuronal circuits during brain development.
即
STDP是一个时序非对称形式的Hebb学习法则,是由突触前和突触后神经元峰值之间的紧密时间相关性影响的。与其他形式的突触可塑性一样,人们普遍认为它是大脑学习和信息存储的基础,也是大脑发育过程中神经元回路的发展和完善的基础。
- 题目:Complementary Memtransistor-Based Multilayer Neural Networks for Online Supervised Learning Through (Anti-)Spike-Timing-Dependent Plasticity
- 时间:2020
- 期刊:TNNLS
- 研究机构:华中科技大学
Contribution
利用互补石墨烯铁电memtransistor实现ReSuMe算法,并应用于多层感知机(MNN)
A C-MT-based synapse design is proposed to implement the SL ReSuMe algorithm in MNNs
实现
包括XOR问题和MNIST数据集
The proposed C-MT-based MNN system is examined by numerical calculation
- 题目:A 4096-Neuron 1M-Synapse 3.8-pJ/SOP Spiking Neural Network With On-Chip STDP Learning and Sparse Weights in 10-nm FinFET CMOS
- 时间:2018
- 期刊:JSSC
- 研究机构:英特尔
Contribution
- 利用了SNN权重的稀疏性降低存储开销
- 片上STDP学习
- 利用NoC实现多播通信
- 近似计算 &下降流量控制优化噪声
实现的SNN操作
- 非监督STDP,玻尔兹曼机,实现图像重建监督
- STDP训练SNN,实现分类
- 利用误差传播训练BNN,实现分类