# US 12,112,255 B1 · System, network and method for selective activation of a computing network > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,112,255 B1 - Document ID: US12112255B1 - Status: Granted - Kind: B1 - Application number: 18/379,743 - Filed: 2023-10-13 - Granted: 2024-10-08 - Earliest claimed: 2023-05-18 - Assignees: Silvretta Research, Inc. - Applicants: Silvretta Research, Inc. - Inventors: Giuseppe G. Nuti - Web page: https://patentgrep.com/patent/US12112255B1 ## CPC - G06N 3/047: Probabilistic or stochastic networks - G06N 3/048: Activation functions - G06N 3/063: using electronic means - G06N 3/08: Learning methods - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Embodiments of the present disclosure implement a stochastic neural network (SNN) where nodes are selectively activated depending on the inputs and which can be trained on multiple objectives. A system can include one or more nodes and one or more synapses, wherein each synapse connects a respective pair of the plurality of nodes. The system can further include one or more processing elements, wherein each of the processing elements is embedded in a respective synapse, and wherein each of the processing elements is adapted to receive an input and generate an output based on the input. The system can be configured to operate such that, upon receipt of a first problem input, a first subset of the nodes in the system is selectively activated. Upon receipt of a second problem input, a second subset of the nodes is selectively activated. The second subset of nodes can be different from the first subset of nodes. In various embodiments, the first and second subsets of nodes can be selectively activated in parallel. ## Claims ### Claim 1 A method for activating a stochastic neural network, comprising: providing a stochastic neural network comprising a plurality of nodes; providing a plurality of synapses, wherein each of the plurality of synapses comprises a connection between a respective pair of the plurality of nodes; providing a plurality of activation weights, wherein each of the plurality of activation weights is associated with a respective synapse of the plurality of synapses; and upon receiving a first problem input and a second problem input, concurrently: activating a first subset of the plurality of nodes based upon the first problem input and the activation weight of one or more of the plurality of synapses that feed into the first subset of the plurality of nodes, and activating a second subset of the plurality of nodes based upon the second problem input and the activation weight of one or more of the plurality of synapses that feed into the second subset of the plurality of nodes. ### Claim 2 (depends on 1) The method of claim 1, wherein the first problem input is different from the second problem input and wherein the first subset of the plurality of nodes is the same as the second subset of the plurality of nodes. ### Claim 3 (depends on 1) The method of claim 1, wherein the first problem input is the same as the second problem input and wherein the first subset of the plurality of nodes is different from the second subset of the plurality of nodes. ### Claim 4 (depends on 1) The method of claim 1, wherein the first problem input is different from the second problem input and wherein the first subset of the plurality of nodes is different from the second subset of the plurality of nodes. ### Claim 5 (depends on 1) The method of claim 1, wherein the first subset of the plurality of nodes comprises a first node, and wherein a computational lock over the first node is selectively applied when activating the first subset of the plurality of nodes. ### Claim 6 (depends on 1) The method of claim 1, wherein a computational lock over one or more of the plurality of synapses is selectively applied. ### Claim 7 (depends on 1) The method of claim 1, wherein at least one of the one or more of the plurality of synapses that feed into the first subset of the plurality of nodes has an output value that has not been computed. ### Claim 8 (depends on 1) The method of claim 1, further comprising: providing a plurality of processing elements, wherein each of the plurality of processing elements is embedded in a respective synapse of the plurality of synapses, wherein each of the plurality of processing elements is adapted to receive an input and generate an output based on the input, wherein the output of a first synapse of the one or more of the plurality of synapses that feed into the first subset of the plurality of nodes comprises a synaptic value computed according to a path preceding the first synapse, and activating a third subset of the plurality of nodes using the synaptic value of the first synapse. ### Claim 9 (depends on 8) The method of claim 8, wherein the path comprises the output of one or more of the plurality of synapses that feed into the first synapse and that are activated. ### Claim 10 (depends on 1) The method of claim 1, further comprising upon determining that first and second synapses of the one or more of the plurality of synapses have correlated inputs into the first node, recycling the first synapse and changing the activation weight for the first synapse. ### Claim 11 (depends on 1) The method of claim 1, wherein the first node of the plurality of nodes feeds into two or more of the plurality of synapses, and wherein the activation weight associated with a first synapse of the two or more of the plurality of synapses is different from the activation weight associated with a second synapse of the two or more of the plurality of synapses. ### Claim 12 (depends on 1) The method of claim 1, wherein the first node of the subset of nodes feeds into a first subset of the plurality of synapses. ### Claim 13 (depends on 1) The method of claim 1, further comprising generating an output based on the first problem input. ### Claim 14 (depends on 13) The method of claim 13, wherein the output comprises a computed value, wherein the method further comprises determining whether the first node of the plurality of nodes will be activated based on the computed value and wherein the computed value comprises an input to a subsequent node of the plurality of nodes after the first node. ### Claim 15 (depends on 1) The method of claim 1, wherein the plurality of nodes comprises a stochastic output node which may or may not be activated. ### Claim 16 A computing system, comprising: a stochastic neural network comprising a plurality of nodes, wherein the plurality of nodes comprises a first subset of nodes and a second subset of nodes, wherein the first subset of nodes comprises a first node; a plurality of synapses, wherein each synapse of the plurality of synapses connects a respective pair of the plurality of nodes; and a plurality of processing elements, wherein each of the plurality of processing elements is embedded in a respective synapse of the plurality of synapses; wherein, upon receipt of a first problem input and a second problem input, the first subset of nodes of the stochastic neural network is activated by at least one of the plurality of synapses and the second subset of nodes of the stochastic neural network is activated by at least one of the plurality of synapses concurrently with the activation of the first subset of nodes. ### Claim 17 (depends on 16) The computing system of claim 16, wherein the first problem input is different from the second problem input and wherein the first subset of the plurality of nodes is the same as the second subset of the plurality of nodes. ### Claim 18 (depends on 16) The computing system of claim 16, wherein the first problem input is the same as the second problem input and wherein the first subset of the plurality of nodes is different from the second subset of the plurality of nodes. ### Claim 19 (depends on 16) The computing system of claim 16, wherein the first problem input is different from the second problem input and wherein the first subset of the plurality of nodes is different from the second subset of the plurality of nodes. ### Claim 20 (depends on 16) The computing system of claim 16, wherein each synapse of the plurality of synapses comprises a signal, a probability of activation and an output, wherein a first synapse of the plurality of synapses feeds into the first node, wherein the first node feeds into a second synapse of the plurality of synapses and wherein the probability of activation of the first synapse is combined with the signal from the second synapse to form a second synapse output. ### Claim 21 (depends on 20) The computing system of claim 20, wherein the output of the first synapse comprises a computed value computed according to a path preceding the first synapse, wherein the computing system determines whether the first node will be activated based on the computed value. ### Claim 22 (depends on 21) The computing system of claim 21, wherein the path comprises the output of one or more of the plurality of synapses that feed into the first synapse and that are activated. ### Claim 23 (depends on 20) The computing system of claim 20, wherein the second synapse output comprises a computed value that comprises an input to a subsequent node of the plurality of nodes after the first node. ### Claim 24 (depends on 16) The computing system of claim 16, wherein each of the plurality of synapses is operable to control activation of at least one of the plurality of nodes. ### Claim 25 (depends on 16) The computing system of claim 16, further comprising a plurality of activation weights, wherein each of the plurality of activation weights is associated with a respective synapse of the plurality of synapses. ### Claim 26 (depends on 25) The computing system of claim 25, wherein activation of the first node of the plurality of nodes is dependent upon the activation weight of the first synapse. ### Claim 27 (depends on 25) The computing system of claim 25, wherein the activation weight associated with the first synapse is different from the activation weight associated with the second synapse. ### Claim 28 (depends on 16) The computing system of claim 16, wherein the plurality of nodes comprises a stochastic output node which may or may not be activated.