发明名称 |
Method and apparatus for neural learning of natural multi-spike trains in spiking neural networks |
摘要 |
Certain aspects of the present disclosure support a technique for neural learning of natural multi-spike trains in spiking neural networks. A synaptic weight can be adapted depending on a resource associated with the synapse, which can be depleted by weight change and can recover over time. In one aspect of the present disclosure, the weight adaptation may depend on a time since the last significant weight change. |
申请公布号 |
US9111224(B2) |
申请公布日期 |
2015.08.18 |
申请号 |
US201113276812 |
申请日期 |
2011.10.19 |
申请人 |
QUALCOMM Incorporated |
发明人 |
Hunzinger Jason Frank |
分类号 |
G06N3/08;G06N3/04 |
主分类号 |
G06N3/08 |
代理机构 |
|
代理人 |
Patel Rupit M. |
主权项 |
1. A method of neural learning, comprising:
modulating an input from a pre-synaptic neuron to a post-synaptic neuron by a weight of a synapse connecting the pre-synaptic neuron and the post-synaptic neuron; and adapting the weight of the synapse based on a resource associated with the synapse, wherein a value of the resource is modified as a function of time relative to adapting the weight of the synapse, such that the value of the resource is recovered over time towards a previous value before the modification. |
地址 |
San Diego CA US |