Parallel Ensemble Machine Learning Algorithm Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Contemporary machine learning algorithms face inefficiencies as data size increases, leading to higher resource utilization and processing time, particularly in training and testing processes, which are not effectively managed by existing methods.
Innovation Solution
The method involves breaking machine learning training data into non-overlapping subsets of features and observations, training multiple algorithms in parallel, and selecting the most accurate one based on validation and testing results, allowing for efficient resource utilization and reduced turnaround time through a parallel configuration and ensemble-based approach.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning algorithms are trained on large datasets, then prediction accuracy is improved, but processing time and resource utilization increase
Solution Approach 1:
The patent divides the large training dataset into multiple non-overlapping subsets, where each subset contains a portion of the features and observations. Multiple machine learning algorithms are trained in parallel on these different subsets simultaneously, allowing the system to process large datasets without proportionally increasing processing time, as the workload is distributed across multiple algorithms operating concurrently.
2Measurement precision
If machine learning algorithms are trained on large datasets, then prediction accuracy is improved, but resource utilization increases
Solution Approach 1:
By segmenting the training data into subsets and distributing them across multiple algorithms that run in parallel, the computational burden on any single algorithm is reduced. This allows the system to leverage large datasets for improved accuracy while distributing resource consumption across multiple processes, preventing any single resource bottleneck and improving overall efficiency.
Solution Approach 2:
The patent combines multiple machine learning algorithms into an ensemble system where each algorithm processes a different subset of the training data. The results from these parallel algorithms are then integrated through testing and selection, allowing the system to achieve high prediction accuracy by leveraging the collective strength of multiple algorithms while distributing resource utilization across the ensemble.
3Productivity
If multiple machine learning algorithms are trained in parallel, then processing time is reduced, but device complexity increases
Solution Approach 1:
The patent segments both the data and the computational tasks by dividing the training dataset into subsets and assigning each subset to a different machine learning algorithm. This segmentation enables parallel processing that reduces training time, while the modular structure of independent algorithms on independent data subsets keeps the system complexity manageable through clear task separation.
Data Source
AI summary
An aspect of the invention includes receiving machine learning (ML) training data that includes a plurality of features for a plurality of observations. The ML training data is broken into a plurality of non-overlapping subsets of features and observations. A first ML algorithm is trained based on a first subset of the features and observations, and a second ML algorithm is trained based on a second subset of the features and observations. The training of the first ML algorithm overlaps in time with the training of the second ML algorithm. The first and second ML algorithms are tested. Either the first or second ML algorithm is selected based at least in part on results of the testing. The selected ML algorithm is retained as a trained ML algorithm for predicting one or more of the plurality of features based on one or more others of the plurality of features.