EP-4742101-A1 - RAPIDLY DEPLOYABLE AGENTIC REASONING PLATFORM

EP4742101A1EP 4742101 A1EP4742101 A1EP 4742101A1EP-4742101-A1

Abstract

Systems, methods, or techniques are provided for reasoning and response generation across a synthetic data mesh. In various embodiments, a system can comprise a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise: a plurality of reasoning components, wherein a first reasoning component of the plurality of reasoning components is operatively coupled to other reasoning components of the plurality of reasoning components, and the first reasoning component is configured to select one or more of the other reasoning components to assist in responding to a prompt.

Inventors

Ninteman III, Lambert Joseph
Bonevich, Jeffrey Donald

Assignees

Fisher Scientific Company, L.L.C.

Dates

Publication Date: 20260513
Application Date: 20251106

Claims (20)

A system for response generation across a network of distributed computing nodes, comprising: a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise: a plurality of reasoning components, wherein a first reasoning component of the plurality of reasoning components is operatively coupled to other reasoning components of the plurality of reasoning components, and the first reasoning component is configured to select one or more of the other reasoning components to invoke to respond to a prompt.
The system of claim 1, wherein the first reasoning component is further configured to select one or more artificial intelligence agents of a plurality of artificial intelligence agents to invoke to respond to the prompt.
The system of claim 2, wherein the first reasoning component is configured to select the one or more other reasoning components of the plurality of reasoning components to invoke to respond to the prompt by calculating a relevance metric, to the prompt, of lists of capabilities of the plurality of reasoning components.
The system of claim 1, wherein one or more reasoning components of the plurality of reasoning components are configured to select one or more additional reasoning components of the plurality of reasoning components based on one or more additional prompts.
The system of claim 3, wherein the first reasoning component is further configured to generate a response to the prompt based on data provided by the selected one or more artificial intelligence agents and the selected one or more other reasoning components.
The system of claim 5, wherein the first reasoning component is configured to generate the response to the prompt by: generating one or more data structures from the data provided by the selected one or more artificial intelligence agents and the selected one or more other reasoning components; generating one or more outputs based on the prompt and the one or more data structures; and synthesizing the response from conclusions of two or more adversarial artificial intelligence agents, wherein the two or more adversarial artificial intelligence agents generate the conclusions based on the one or more outputs.
The system of claim 6, wherein the first reasoning component is further configured to generate the response to the prompt by dimensionally reducing the one or more data structures using a singular value decomposition.
The system of claim 5, wherein a second reasoning component of the selected one or more other reasoning components is configured to, in response to being selected: check at least one of role-based control clearance or attribute-based access control clearance of the first reasoning component; approve data from one or more data sources managed by the second reasoning component based on the at least one of the role-based control clearance or attribute-based access control clearance of the first reasoning component; aggregate relevant data of the approved data based on the prompt; and transmit the aggregated relevant data to the first reasoning component.
A computer-implemented method for response generation across a network of distributed computing nodes comprising: selecting, by a device operatively coupled to a processor, one or more reasoning components of a plurality of operatively coupled reasoning components based on a prompt; selecting, by the device, one or more artificial intelligence agents of a plurality of artificial intelligence agents based on the prompt; and generating, by the device, a response to the prompt based on data generated by the selected one or more reasoning components and data generated by the selected one or more artificial intelligence agents.
The computer-implemented method of claim 9, wherein the one or more selected reasoning components further select one or more additional reasoning components of the plurality of operatively coupled reasoning components.
The computer-implemented method of claim 9, wherein the selecting the one or more artificial intelligence agents of the plurality of artificial intelligence agents comprises calculating, by the device, a relevance metric, to the prompt, of lists of capabilities of the plurality of artificial intelligence agents.
The computer-implemented method of claim 9, wherein the selecting the one or more reasoning components comprises calculating, by the device, a relevance metric, to the prompt, of lists of data sources managed by the plurality of reasoning components.
The computer-implemented method of claim 10, wherein generating the response to the prompt comprises: generating, by the device, using the selected one or more reasoning components, data from one or more data sources managed by the selected one or more reasoning components; populating, by the device, one or more knowledge graphs with the data generated by the selected one or more artificial intelligence agents and the data generated by the selected one or more reasoning components; generating, by the device, using one or more analytical modules of the selected one or more reasoning components, one or more data transformations based on the prompt and the one or more knowledge graphs; and synthesizing, by the device, using natural language processing, the response from arguments of a plurality of adversarial artificial intelligence agents, wherein the plurality of adversarial artificial intelligence agents generate the arguments based on the one or more data transformations.
A computer program product for response generation across a network of distributed computing nodes comprising a non-transitory computer-readable memory having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to: generate, by the processor, using a first reasoning component of a plurality of operatively coupled reasoning components, one or more sub-tasks based on a received prompt; select, by the processor, using the first reasoning component, one or more additional reasoning components of the plurality of reasoning components to execute the one or more sub-tasks; execute, by the processor, using the one or more additional reasoning components, the one or more sub-tasks; transmit, by the processor, using the one or more additional reasoning components, results of the one or more sub-tasks to the first reasoning component; and generate, by the processor, using the first reasoning component, a response to the received prompt based on the results of the one or more sub-tasks.
The computer program product of claim 14, wherein the transmitting the results of the one or more sub-tasks comprises a publish/subscriber communication protocol between the first reasoning component and the selected one or more additional reasoning components.
The computer program product of claim 14, wherein the program instructions are further executable to cause the processor to, in response to a reasoning component of the one or more additional reasoning components being selected to execute a sub-task of the one or more sub-tasks: determine, by the processor, using the reasoning component, if the sub-task complies with security policies of the reasoning component; and in response to determining the sub-task complies with the one or more security policies of the reasoning component, execute, by the processor, using the reasoning component, the sub-task.
The computer program product of claim 14, wherein the program instructions are further executable to cause the processor to: determine, by the processor, using the first reasoning component, if the one or more sub-tasks comply with security policies of the first reasoning component; and in response to determining the one or more sub-tasks comply with the security policies, select, by the processor, using the first reasoning component, the one or more additional reasoning components of the plurality of reasoning components to execute the one or more sub-tasks.
The computer program product of claim 14, wherein the program instructions are further executable to cause the processor to: receive, by the processor, using the first reasoning component, the received prompt; determine, by the processor, using the first reasoning component, if the received prompt complies with security policies of the first reasoning component; and in response to determining the received prompt complies with the security policies of the first reasoning component, generate, by the processor, using the first reasoning component, the one or more sub-tasks based on the received prompt.
The computer program product of claim 14, wherein the processing instructions are further executable by the processor to cause the processor to: determine, by the processor, using the first reasoning component, if the response to the prompt complies with security policies of the first reasoning component; and in response to determining the response to the prompt complies with the security policies of the first reasoning component, display, by the processor, the response to the received prompt on a graphical user interface.
The computer program product of claim 14, wherein the selecting of the one or more additional reasoning components to execute the sub-tasks causes the processor to: determine, by the processor, using the first reasoning component, a database comprising data relevant to a sub-task of the one or more sub-tasks; identify, by the processor, using the first reasoning component, a reasoning component of the plurality of reasoning components that manages the database; and transmitting, by the processor, using the first reasoning component, the sub-task to the identified reasoning component.

Description

Cross-Reference to Related Application This application claims priority to and the benefit of U.S. Provisional Application No. 63/718,094, entitled "RAPIDLY DEPLOYABLE AGENTIC REASONING PLATFORM," which was filed on November 8, 2024. The aforementioned application is hereby incorporated herein by reference in its entirety. Background Many industries and enterprises are rapidly introducing artificial intelligence (Al) into their workflows and processes. However, integrating various data sources as well as providing enterprise-wide access to AI processes can lead to significant scaling issues. Summary The following presents a summary to provide a basic understanding of one or more embodiments. This summary is not intended to identify key or critical elements, or delineate any scope of the particular embodiments or any scope of the claims. Its sole purpose is to present concepts in a simplified form as a prelude to the more detailed description that is presented later. In one or more embodiments described herein, devices, systems, methods, or apparatus that facilitate reasoning and response generation across a synthetic data mesh are provided. According to one or more embodiments, a system is provided. The system can comprise a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise: a plurality of reasoning components, wherein a first reasoning component of the plurality of reasoning components is operatively coupled to other reasoning components of the plurality of reasoning components, and the first reasoning component is configured to select one or more of the other reasoning components to assist in responding to a prompt. An advantage of the system, and/or of a corresponding method, can be rapid and easy scaling of the system through the use and addition of new reasoning components to the plurality of reasoning components. Brief Description of the Drawings Embodiments will be readily understood by the following detailed description in conjunction with the accompanying drawings. To facilitate this description, like reference numerals designate like structural elements. Embodiments are illustrated by way of example, not by way of limitation, in the figures of the accompanying drawings. FIG. 1 is a block diagram of an example scientific instrument module for performing reasoning and response generation across a synthetic data mesh, in accordance with one or more embodiments described herein.FIG. 2 is a flow diagram of an example method of performing reasoning and response generation across a synthetic data mesh, in accordance with one or more embodiments described herein.FIG. 3 illustrates a block diagram of an example system that can facilitate reasoning and response generation across a synthetic data mesh, in accordance with one or more embodiments described herein.FIG. 4 illustrates a block diagram of an example reasoning component, in accordance with one or more embodiments described herein.FIG. 5 illustrates a block diagram of an example reasoning engine, in accordance with one or more embodiments described herein.FIG. 6 illustrates a block diagram of an example knowledge engine, in accordance with one or more embodiments described herein.FIG. 7 illustrates a block diagram of example analytical modules, in accordance with one or more embodiments described herein.FIG. 8 illustrates a flow diagram of an example method that can facilitate reasoning and response generation across a plurality of reasoning components, in accordance with one or more embodiments described herein.FIG. 9 illustrates a flow diagram of an example method that can facilitate response generation across a plurality of reasoning components, in accordance with one or more embodiments described herein.FIG. 10 illustrates a flow diagram of an example method that can facilitate information transfer across a plurality of reasoning components, in accordance with one or more embodiments described herein.FIG. 11 illustrates a block diagram of an example operating environment in which one or more embodiments described herein can be facilitated. Detailed Description The following detailed description is merely illustrative and is not intended to limit embodiments and/or application or utilization of embodiments. One or more embodiments are now described with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a more thorough understanding of the one or more embodiments. It is evident, however, in various cases, that the one or more embodiments can be practiced without these specific details. In various fields and enterprises, AI is increasingly being applied across large networks and to large existing data sources. Often, these AI applications are hosted centrally with users accessing the AI application remotely. However,