# US 11,907,847 B2 · Operating machine-learning models on different platforms > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,907,847 B2 - Document ID: US11907847B2 - Status: Granted - Kind: B2 - Application number: 17/182,254 - Filed: 2021-02-23 - Published: 2021-06-10 - Granted: 2024-02-20 - Earliest claimed: 2016-06-30 - Assignees: Cogniac, Corp. - Applicants: Cogniac, Corp. - Inventors: William S Kish; Huayan Wang; Sandip C. Patel - Web page: https://patentgrep.com/patent/US11907847B2 ## CPC - G06N 3/084: Backpropagation, e.g. using gradient descent - G06N 3/063: using electronic means - G06N 20/00: Machine learning - G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An electronic device may determine whether a machine-learning model is operating within predefined limits. In particular, the electronic device may receive, from another electronic device, instructions for the machine-learning model, a reference input and a predetermined output of the machine-learning model for the reference input. Note that the instructions may include an architecture of the machine-learning model, weights associated with the machine-learning model and/or a set of pre-processing transformations for use when executing the machine-learning model on images. In response, the electronic device may configure the machine-learning model based on the instructions. Then, the electronic device may calculate an output of the machine-learning model for the reference input. Next, the electronic device may determine whether the machine-learning model is operating within predefined limits based on the output and the predetermined output. ## Claims ### Claim 1 An electronic device, comprising: an interface circuit configured to communicate, via a network, with a second electronic device; memory configured to store program instructions; and a processor coupled to the interface circuit and the memory, wherein, when executed by the processor, the program instructions cause the electronic device to perform operations comprising: receiving, from the second electronic device, instructions for a machine-learning model, a reference input and a predetermined output of the machine-learning model for the reference input, wherein the instructions include an architecture of the machine-learning model; and determining whether the machine-learning model is operating on the electronic device within predefined limits on an accuracy of the machine-learning model based at least in part on a calculated output of the machine-learning model for the reference input and the predetermined output, wherein the predefined limits assess whether the machine-learning model is operating correctly on the electronic device based at least in part on the electronic device having at least one of a different processor, a different operating system or a different version of the operating system than a computer or a computer system that trained the machine-learning model and that is different from the electronic device. ### Claim 2 (depends on 1) The electronic device of claim 1, wherein the machine-learning model includes a neural network. ### Claim 3 (depends on 1) The electronic device of claim 1, wherein the operations further comprise, when the machine-learning model is operating outside of the predefined limits, performing a remedial action. ### Claim 4 (depends on 1) The electronic device of claim 1, wherein the operations further comprise: when the machine-learning model is operating within the predefined limits, analyzing images using the machine-learning model. ### Claim 5 (depends on 4) The electronic device of claim 4, wherein the operations further comprise: prior to the analysis, mapping the images to a set of input images having associated formats that are compatible with the machine-learning model using one or more of a set of pre-processing transformations that are associated with the machine-learning model. ### Claim 6 (depends on 5) The electronic device of claim 5, wherein a given pre-processing transformation is associated with at least a subset of the images based on an image context of at least the subset of the images; and wherein the image context includes at least one of: a format of at least the subset of the images, and an image characteristic associated with at least the subset of the images. ### Claim 7 (depends on 6) The electronic device of claim 6, wherein the instructions specify the image context, so that the given pre-processing transformation is applied to the subset of the images prior to the analysis using the machine-learning model. ### Claim 8 (depends on 6) The electronic device of claim 6, wherein the operations further comprise: determining the image context after receiving the instructions, so that the given pre-processing transformation is applied to the subset of the images prior to the analysis using the machine-learning model. ### Claim 9 (depends on 8) The electronic device of claim 8, wherein the determination of the image context is performed dynamically to improve a performance metric of the machine-learning model. ### Claim 10 (depends on 1) The electronic device of claim 1, wherein the operations further comprise: when the machine-learning model is operating outside of the predefined limits, providing a notification to the second electronic device and discontinuing analysis of images using the machine-learning model. ### Claim 11 (depends on 1) The electronic device of claim 1, wherein the reference input comprises an image pre-processed using one of a set of pre-processing transformations that are associated with the machine-learning model. ### Claim 12 A non-transitory computer-readable storage medium for use in conjunction with an electronic device, the computer-readable storage medium configured to store program instructions that, when executed by the electronic device, causes the electronic device to determine whether a machine-learning model is operating within predefined limits by performing operations comprising: receiving, from a second electronic device, instructions for the machine-learning model, a reference input and a predetermined output of the machine-learning model for the reference input, wherein the instructions include an architecture of the machine-learning model; and determining whether the machine-learning model is operating on the electronic device within predefined limits on an accuracy of the machine-learning model based at least in part on a calculated output of the machine-learning model for the reference input and the predetermined output, wherein the predefined limits assess whether the machine-learning model is operating correctly on the electronic device based at least in part on the electronic device having at least one of a different processor, a different operating system or a different version of the operating system than a computer or a computer system that trained the machine-learning model that is different from the electronic device. ### Claim 13 (depends on 12) The computer-readable storage medium of claim 12, wherein the machine-learning model includes a neural network. ### Claim 14 (depends on 12) The computer-readable storage medium of claim 12, wherein the operations further comprise, when the machine-learning model is operating outside of the predefined limits, performing a remedial action. ### Claim 15 (depends on 12) The computer-readable storage medium of claim 12, wherein the operations further comprise: when the machine-learning model is operating within the predefined limits, analyzing images using the machine-learning model. ### Claim 16 (depends on 15) The computer-readable storage medium of claim 15, wherein the operations further comprise: prior to the analysis, mapping the images to a set of input images having associated formats that are compatible with the machine-learning model using one or more of a set of pre-processing transformations that are associated with the machine-learning model. ### Claim 17 (depends on 16) The computer-readable storage medium of claim 16, wherein a given pre-processing transformation is associated with at least a subset of the images based on an image context of at least the subset of the images; and wherein the image context includes at least one of: a format of at least the subset of the images, and an image characteristic associated with at least the subset of the images. ### Claim 18 (depends on 17) The computer-readable storage medium of claim 17, wherein the instructions specify the image context, so that the given pre-processing transformation is applied to the subset of the images prior to the analysis using the machine-learning model. ### Claim 19 (depends on 17) The computer-readable storage medium of claim 17, wherein the operations further comprise: determining the image context after receiving the instructions, so that the given pre-processing transformation is applied to the subset of the images prior to the analysis using the machine-learning model. ### Claim 20 A method for determining whether a machine-learning model is operating within predefined limits, the method comprising: by an electronic device: receiving, from a second electronic device, instructions for the machine-learning model, a reference input and a predetermined output of the machine-learning model for the reference input, wherein the instructions include an architecture of the machine-learning model; and determining whether the machine-learning model is operating on the electronic device within predefined limits on an accuracy of the machine-learning model based at least in part on a calculated output of the machine-learning model for the reference input and the predetermined output, wherein the predefined limits assess whether the machine-learning model is operating correctly on the electronic device based at least in part on the electronic device having at least one of a different processor, a different operating system or a different version of the operating system than a computer or a computer system that trained the machine-learning model that is different from the electronic device.