Machine Learning Model Ensemble for Classification Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing classification techniques using variational autoencoders (VAEs) require creating a VAE for each class, leading to increased computational time due to the complexity of the process.
Innovation Solution
A method employing a vector neural network with multiple machine learning models, each configured to classify input data into multiple classes, uses a similarity calculation between feature vectors to determine the class of classification target data, allowing for efficient classification processing by selecting the appropriate model from a set of N machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a VAE is created for each class to perform classification, then classification accuracy is improved, but computational time increases significantly
Solution Approach 1:
The patent divides the classification task into multiple specialized machine learning models, where each model is trained to classify into a specific subset of classes rather than all classes. This segmentation allows each model to be more efficient while maintaining overall classification accuracy across all classes through the ensemble of specialized models.
Solution Approach 2:
Each machine learning model in the patent is designed with multi-functionality to handle multiple classes within its specialized domain. The models use shared feature extraction layers that can process various types of input data, making them versatile while still being specialized for their specific class subsets, thus improving both accuracy and computational efficiency.
2Measurement precision
If multiple machine learning models are prepared with overlapping class configurations, then classification accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the overall classification problem into multiple smaller classification problems, each handled by a specialized machine learning model. Each model is configured to classify into a specific subset of classes, which reduces the complexity of individual models while maintaining high accuracy through the collective capability of all models.
Solution Approach 2:
The system dynamically selects which machine learning model to use based on the input data characteristics and the required class classification. This dynamic selection approach allows the system to adapt to different classification scenarios, reducing the effective complexity by only activating the necessary model for each specific classification task.
Data Source
AI summary
A method of performing classification processing on classification target data includes: (a) a step of preparing N machine learning models; (b) a step of, when a plurality of pieces of training data are input into the N machine learning models, preparing a known feature vector group obtained from output of at least one specific layer of the plurality of vector neuron layers; and (c) a step of computing, using a selected machine learning model selected from the N machine learning models a similarity, for each class, between the known feature vector group and a feature vector obtained from output of the specific layer when the classification target data is input into the selected machine learning model, and determining a class for the classification target data using the similarity.