Planning and Task Agents for Complex Computing Tasks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of computing platforms often face challenges in defining and executing complex computing tasks without explicit knowledge of the underlying basic operations, leading to time-consuming and cumbersome processes.
Innovation Solution
A computer-implemented method using planning and task agents, coupled with large language models, to decompose high-level use case inputs into executable tasks, identify suitable tools, and execute operations in the computing platform, while managing dependencies and user input requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If users directly define complex computing tasks using basic operations, then execution precision is improved, but ease of operation deteriorates
Solution Approach 1:
The system segments complex computing tasks into smaller sub-tasks and individual operations. The planning agent breaks down high-level objectives into discrete steps, while task agents further decompose these into basic operations that can be executed by the computing platform. This segmentation allows users to work at a high level without needing to understand underlying complexity.
Solution Approach 2:
The patent introduces agent systems as intermediaries between users and the computing platform. Planning agents translate user objectives into task sequences, while task agents convert tasks into executable operations. These intermediary agents bridge the gap between high-level user intent and low-level platform operations, eliminating the need for users to directly define complex sequences of basic operations.
2Reliability
If users explicitly define all basic operations, then task execution reliability is improved, but time consumption increases
Solution Approach 1:
The system performs preliminary actions by pre-defining operation libraries with standardized basic operations and their relationships. Task templates and operation schemas are prepared in advance, allowing the agent system to quickly assemble reliable execution sequences without users needing to define everything from scratch. This preliminary preparation maintains reliability while reducing time consumption.
3Ease of operation
If the system provides high-level functionality abstraction, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The agent system is designed as a universal intermediary that can handle multiple types of computing tasks and objectives. The planning agent and task agents can work with various operation types and platform functionalities through standardized interfaces. This universality allows high-level abstraction for ease of operation while managing complexity through a unified agent architecture rather than separate mechanisms for each task type.
Data Source
AI summary
Methods, apparatuses and computer programs are for executing complex computing tasks in a computing platform are provided. According to one aspect, a method comprises receiving, by a planning agent, a use case input indicating an objective for completion in the computing platform. The planning agent decomposes, by the planning agent, the use case input into a plurality of tasks for achieving the objective. The planning agent provides each of the plurality of tasks to a respective task agent for execution. For each task of the plurality of tasks, the respective task agent identifies a tool suitable for performing the task from a plurality of tools. The task agent uses the identified tool to execute an operation corresponding to the respective task in the computing platform.