Universal Task Learning System for Cross-Application Execution

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional personal assistants and smart agents require redundant implementation of specific application programming interfaces (APIs) for each application to perform tasks, leading to inefficiencies and increased development efforts across multiple applications.

Innovation Solution

The Semantic Intelligent Task Learning and Adaptive Execution System (STaLES) learns tasks from one application and applies them across related applications with minimal training, using AI processing to capture and understand user interactions, visual data, and natural language, enabling dynamic adaptation and execution on different applications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional personal assistants implement specific APIs for each application to perform tasks, then task execution reliability is improved, but device complexity and development effort increase

Engineering Contradiction:

Improvetask execution reliabilityVSAvoiddevelopment complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements a universal task learning system that can execute tasks across multiple applications using a single learned task model. The system captures features and data from interactions with one application and generalizes the learned task to execute on related applications without requiring separate API implementations for each app, thus reducing development complexity while maintaining execution reliability

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system creates a copied representation of task execution patterns from one application and applies this copied knowledge to other applications. By capturing and storing task features, interaction data, and execution sequences as reusable templates, the system eliminates the need to re-implement tasks for each application individually

Inventive Principle:

Principle #26Copying

2Adaptability or versatility

If redundant API implementations are created for each application, then task adaptability across applications is improved, but loss of time and development resources increase

Engineering Contradiction:

Improvetask adaptabilityVSAvoiddevelopment time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary task learning by capturing features and interaction data during initial demonstrations with one application. This preliminary action creates a reusable task model that can be quickly adapted to related applications, eliminating the need for time-consuming redundant implementations while maintaining broad adaptability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent merges task execution logic across multiple applications by consolidating common task patterns into a single learned model. By combining feature capture, data collection, and task execution into an integrated system that works across application boundaries, it achieves broad adaptability without the time cost of separate implementations

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS11468881B2Method and system for semantic intelligent task learning and adaptive execution

Publication Date: 2022.10.11 SAMSUNG ELECTRONICS CO LTD

AI summary

A method includes receiving, at an electronic device, a command directed to a first application operated by the electronic device. Features presented by the first application in response to interactions with the first application are captured at the electronic device. Data communicated with the first application via the interactions with the first application are captured at the electronic device. A task is learned based on the captured features and communicated data.