# US 2018/0025279 A1 · Cognitive computing for servers and mobile devices > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2018/0025279 A1 - Document ID: US20180025279A1 - Status: Published application - Kind: A1 - Application number: 15/214,092 - Filed: 2016-07-19 - Published: 2018-01-25 - Earliest claimed: 2016-07-19 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Pradip Bose; Alper Buyuktosunoglu; Augusto Javier Vega - Web page: https://patentgrep.com/patent/US20180025279A1 ## CPC - G06N 20/00: Machine learning - G06N 5/04: Inference or reasoning models - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Techniques for decoupling cognitive model training from execution of the cognitive model are provided. In one example, a computer program product is provided that determines cognitive data based on context data and a model of interpreting the context data. The cognitive data can comprise prediction data that represents a prediction relating to a state of an environment. The context data and the cognitive data can be transmitted to a server, and an updated model can be received in response. ## Claims ### Claim 1 A computer program product for interpreting context information, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to: determine, by the processor, cognitive data comprising prediction data that represents a prediction relating to a state of an environment, wherein the determination is based on context data that represents the state of the environment and based on first model data that describes a machine-learned model of interpreting the context data; facilitate, by the processor, transmission of at least a portion of the context data and at least a portion of the cognitive data to a first device; and receive, by the processor, update data comprising second model data that represents an update to the first model data, wherein reception is in response to the transmission. ### Claim 2 (depends on 1) The computer program product of claim 1, wherein the program instructions are further executable by the processor to cause the processor to determine, by the processor, confidence data that represents a confidence measure of the prediction, wherein the cognitive data comprises the confidence data. ### Claim 3 (depends on 1) The computer program product of claim 1, wherein the program instructions are further executable by the processor to cause the processor to determine, by the processor, action data that describes an action to be performed in response to the prediction, wherein the cognitive data comprises the action data, and wherein training associated with the second model data is decoupled from execution of the second model data. ### Claim 4 (depends on 1) The computer program product of claim 1, wherein the update data further comprises updated action data that describes an updated action to be performed in response to the prediction. ### Claim 5 (depends on 1) The computer program product of claim 1, wherein the program instructions are further executable by the processor to cause the processor to: determine, by the processor, third model data based on a machine-learning analysis of the context data; and update, by the processor, the first model data or the second model data with the third model data. ### Claim 6 (depends on 5) The computer program product of claim 5, wherein the program instructions are further executable by the processor to cause the processor to: facilitate, by the processor, transmission of the third model data to the first device. ### Claim 7 (depends on 5) The computer program product of claim 5, wherein the program instructions are further executable by the processor to cause the processor to: facilitate, by the processor, transmission of the third model data or the second model data to a second device that is registered with the first device. ### Claim 8 (depends on 1) The computer program product of claim 1, wherein the program instructions are further executable by the processor to cause the processor to: receive, by the processor, the first model data from the first device or a second device that is registered with the first device, wherein reception of the first model data is based on device type data. ### Claim 9 (depends on 1) The computer program product of claim 1, wherein the program instructions are further executable by the processor to cause the processor to: generate, by the processor, the context data based on sensor data received from a sensor device that monitors the environment. ### Claim 10 (depends on 9) The computer program product of claim 9, wherein the sensor device comprises a camera device or other imaging device. ### Claim 11 (depends on 9) The computer program product of claim 9, wherein the sensor device comprises a motion-sensing device that determines motion-based measurements of the sensor device or of an object in the environment. ### Claim 12 A system, comprising: a memory that stores computer executable components; a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise: a communication component that receives, from a first device, context data that represents extrinsic data monitored by a sensor device; and a cognitive server component that, based on a machine-learning analysis of the context data, determines first model data that represents an artificial intelligence-based model trained to interpret the context data, wherein the communication component transmits the first model data to the first device. ### Claim 13 (depends on 12) The system of claim 12, wherein the communication component further receives, from the first device, cognitive data comprising: prediction data, representing a prediction relating to the extrinsic data, determined based on second model data that is distinct from the first model data; confidence data, representing a determined confidence metric of the prediction, determined based on the second model data; and action data representing an action to be performed in response to the prediction. ### Claim 14 (depends on 13) The system of claim 13, wherein the cognitive server component further determines updated action data representing an updated action to be performed in response to the prediction, and the communication component further transmits the updated action data to the first device. ### Claim 15 (depends on 13) The system of claim 13, wherein the cognitive server component further determines updated action data representing an updated action to be performed in response to an updated prediction determined based on the context data and the first model data, and the communication component further transmits the updated action data to the first device. ### Claim 16 (depends on 12) The system of claim 12, wherein the cognitive server component further determines third model data representing an update to the first model data generated based on the context data, wherein training associated with the third model data is decoupled from execution of the third model data. ### Claim 17 A computer-implemented method, comprising: determining, by a device operatively coupled to a processor, cognitive data comprising prediction data that represents a prediction relating to a characteristic of an environment based on context data that represents the characteristic of the environment and based on first model data that describes a model of interpreting the context data; and receiving, by the device, update data comprising second model data that represents an update to the first model data, wherein the receiving is based on transmission of at least a portion of the context data and at least a portion of the cognitive data to a first device. ### Claim 18 (depends on 17) The computer-implemented method of claim 17, further comprising determining, by the device, confidence data that represents a confidence value associated with the prediction. ### Claim 19 (depends on 17) The computer-implemented method of claim 17, further comprising determining, by the device, action data that describes an action to be performed in view of the prediction. ### Claim 20 (depends on 17) The computer-implemented method of claim 17, further comprising receiving, by the device, third model data representing an update to the second model data in response to a determination that the second model data is outdated. ### Claim at position 21 21-30. (canceled)