Boab
I wish I could paint like Vincent
Unlike @Bravo's hot tub party it looks like our asses are well and truly covered.Hi Boab,
This is the link to the Espacennet publication.
US11657257B2 Spiking neural network
The specification and claims are not available here yet. Perhaps on Google patents?
The "Published As" data shows where it was filed ( the ones with a "B" suffix are granted) but this is complicated as this seems to be a Continuation of the earlier US11468299, probably based on US2023026363. The AU2021254524B2 is probably the parent, not the continuation (divisional /additional in AU).
US2023026363A1 SPIKING NEURAL NETWORK
[0002] This application is a continuation of U.S. patent application Ser. No. 16/670,368, filed on Oct. 31, 2019, which claims the benefit of U.S. Provisional Application No. 62/754,348, filed on Nov. 1, 2018, titled “An Improved Spiking Neural Network,” all of which are hereby incorporated by reference in their entirety for all purposes.
This is the first claim of the application. It may have been modified during examination.
Claim 1
A neuromorphic integrated circuit, comprising:
a spike converter configured to generate spikes from input data;
a spike input buffer configured to store the spikes from the spike converter;
a neuron fabric comprising a neural processor comprising a plurality of spiking neuron circuits configured to perform a task based on the spikes and a neural network configuration;
a memory comprising the neural network configuration, wherein the neural network configuration comprises a potential array and a plurality of synapses, the neural network configuration defines connections between the plurality of spiking neuron circuits and the plurality of synapses, the potential array comprising membrane potential values for the plurality of spiking neuron circuits, and the plurality of synapses having corresponding synaptic weights;
a packetizer configured to generate a spike packet comprising spike bits representing the spikes in the spike input buffer, wherein each spike bit in the spike packet corresponds to a synapse in the plurality of synapses; and
a processor configured to modify the neural network configuration based on a plurality of configuration parameters,
wherein a spiking neuron circuit in the plurality of spiking neuron circuits is configured to:
apply a first logical AND function to a first spike bit in the spike packet and a first synaptic weight of a first synapse corresponding to the first spike bit, wherein the first synaptic weight has a value of one, and the applying the first logical AND function outputs a logical one; and
increment a membrane potential value associated with the spiking neuron circuit in response to the applying the first logical AND function outputting the logical one;
apply a second logical AND function to the first spike bit in the spike packet and a second synaptic weight corresponding to the first spike bit, wherein the second synaptic weight has a negative value, and the applying the second logical AND function outputs a logical one;
decrement the membrane potential value associated with the spiking neuron circuit in response to the applying the second logical AND function outputting the logical one.
For comparison, this is claim 1 of the parent US11468299 which we have seen before:
1. A neuromorphic integrated circuit, comprising:
a spike converter circuit configured to generate spikes from input data;
a reconfigurable neuron fabric comprising a neural processor comprising a plurality of spiking neuron circuits configured to perform a task based on the spikes and a neural network configuration; and
a memory comprising the neural network configuration, wherein the neural network configuration comprises a potential array and a plurality of synapses, and the neural network configuration defines connections between the plurality of spiking neuron circuits and the plurality of synapses, the potential array comprising membrane potential values for the plurality of spiking neuron circuits, and the plurality of synapses having corresponding synaptic weights,
wherein the neural processor is configured to:
select a spiking neuron circuit in the plurality of spiking neuron circuits based on the selected spiking neuron circuit having a membrane potential value that is a highest value among the membrane potential values for the plurality of spiking neuron circuits;
determine that the membrane potential value of the selected spiking neuron circuit reached a learning threshold value associated with the selected spiking neuron circuit; and
perform a Spike Time Dependent Plasticity (STDP) learning function based on the determination that the membrane potential value of the selected spiking neuron circuit reached the learning threshold value associated with the selected spiking neuron circuit.
Thoroughly thorough.
Thank you