InFusionLayer: a CFA-based ensemble tool to generate new classifiers for learning and modeling

arXiv cs.AI / 3/12/2026

📰 NewsTools & Practical UsageModels & Research

共有:

Key Points

InFusionLayer is a machine learning architecture based on Combinatorial Fusion Analysis (CFA) that generates new classifiers by fusing a moderate set of base models for both unsupervised and supervised multiclassification tasks.
The approach leverages CFA concepts such as rank-score characteristics (RSC) and cognitive diversity (CD) to enhance ensemble performance and model fusion.
The project emphasizes practical usability across PyTorch, TensorFlow, and Scikit-learn workflows and demonstrates competitive results on computer vision datasets.
The authors have open-sourced the code on GitHub to encourage community-driven development and broader adoption of CFA in ensemble learning.

Abstract

Ensemble learning is a well established body of methods for machine learning to enhance predictive performance by combining multiple algorithms/models. Combinatorial Fusion Analysis (CFA) has provided method and practice for combining multiple scoring systems, using rank-score characteristic (RSC) function and cognitive diversity (CD), including ensemble method and model fusion. However, there is no general-purpose Python tool available that incorporate these techniques. In this paper we introduce \texttt{InFusionLayer}, a machine learning architecture inspired by CFA at the system fusion level that uses a moderate set of base models to optimize unsupervised and supervised learning multiclassification problems. We demonstrate \texttt{InFusionLayer}'s ease of use for PyTorch, TensorFlow, and Scikit-learn workflows by validating its performance on various computer vision datasets. Our results highlight the practical advantages of incorporating distinctive features of RSC function and CD, paving the way for more sophisticated ensemble learning applications in machine learning. We open-sourced our code to encourage continuing development and community accessibility to leverage CFA on github: https://github.com/ewroginek/Infusion