Edge ML Training Queues Based on Data Confidence Scores

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing edge-assisted machine learning solutions do not account for varying confidence levels of data when training and updating machine learning models, affecting model performance.

Innovation Solution

A DCF-enhanced edge-assisted machine learning system that prioritizes using high confidence data for training and updating machine learning models by segregating data into priority and normal queues based on confidence scores, ensuring higher confidence data is used first.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If all data from edge nodes is used for training machine learning models without prioritization, then the quantity of training data is maximized, but the accuracy and reliability of model training deteriorates due to inclusion of low confidence data

Engineering Contradiction:

Improvequantity of training dataVSAvoidaccuracy of model training

Core Design Contradiction:

Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent segments training data into priority and normal queues based on confidence scores. High confidence data is placed in the priority queue while lower confidence data goes to the normal queue, allowing the system to process data in segments of varying quality for training the machine learning model

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by treating different data points with different levels of importance based on their confidence scores. Rather than uniformly processing all data, the system selectively prioritizes high confidence data for training, assigning different quality weights to different data segments

Inventive Principle:

Principle #3Local quality

2Reliability

If trust services or functions are applied to data to improve confidence levels, then the reliability of data usage is improved, but the cost of providing these services increases

Engineering Contradiction:

Improveconfidence in data usageVSAvoidcost of providing trust services

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent applies partial action by selectively applying trust services only to data that requires enhanced confidence validation, rather than uniformly processing all data. The system identifies and applies trust functions only where necessary to achieve sufficient confidence levels for model training

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The system enables self-service by allowing edge nodes to self-assess and report confidence scores for their generated data. This self-reporting mechanism reduces the need for external trust verification services, as the data sources themselves provide confidence metadata

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20260024005A1DCF-enhanced edge-assisted machine learning

Publication Date: 2026.01.22 DELL PROD LP

AI summary

A computing system may identify data provided by a plurality of edge nodes that is to be used in training or updating a machine learning (ML) model. The computing system may identify a confidence score associated with the data provided by a plurality of edge nodes. The computing system may place the data provided by the plurality of edge nodes that is determined to have a high confidence score into a first queue. The computing system may place the data provided by the plurality of edge nodes that is determined to have a low confidence score into a second queue. The computing system may use the data placed into the first queue and/or the second queue to train or update the ML model.