Autoselection of the Ensemble of Convolutional Neural Networks with Second-Order Cone Programming

Buse Çisil Güldoğuş, Abdullah Nazhat Abdullah, Muhammad Ammar Ali, Süreyya Özöğür-Akyüz
2/12/2023
cs.LGmath.OC

No official code found. Best implementation: Abdullah-88/Ensemble_Autoselection_SOCP (1 star, BSD-3-Clause, last commit 2025-05-10).

Abstract

Ensemble techniques are frequently encountered in machine learning and engineering problems since the method combines different models and produces an optimal predictive solution. The ensemble concept can be adapted to deep learning models to provide robustness and reliability. Due to the growth of the models in deep learning, using ensemble pruning is highly important to deal with computational complexity. Hence, this study proposes a mathematical model which prunes the ensemble of Convolutional Neural Networks (CNN) consisting of different depths and layers that maximizes accuracy and diversity simultaneously with a sparse second order conic optimization model. The proposed model is tested on CIFAR-10, CIFAR-100 and MNIST data sets which gives promising results while reducing the complexity of models, significantly.

View on arXivView PDF

Code Implementations(1)

Abdullah-88/Ensemble_Autoselection_SOCPCommunity100%
BSD-3-Clause
10Python, Jupyter NotebookFeb 20, 20231 years agoBSD-3-Clause

Similar Papers

Enhancing CNN Classification with Lamarckian Memetic Algorithms and Local Search

82% match2024

Sparse-Input Neural Network using Group Concave Regularization

80% match1 repo2023

Hierarchical Pruning of Deep Ensembles with Focal Diversity

80% match1 repo52023

Efficient Latency-Aware CNN Depth Compression via Two-Stage Dynamic Programming

79% match1 repo132023

Layer-Specific Optimization: Sensitivity Based Convolution Layers Basis Search

79% match2024

Trusting SVM for Piecewise Linear CNNs

79% match1 repo82016

Cite this paper

@article{gldou2023autoselection,
  title  = {Autoselection of the Ensemble of Convolutional Neural Networks with Second-Order Cone Programming},
  author = {Buse Çisil Güldoğuş and Abdullah Nazhat Abdullah and Muhammad Ammar Ali and Süreyya Özöğür-Akyüz},
  year   = {2023},
  eprint = {2302.05950},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2302.05950v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2302.05950.svg)](https://codeofpaper.com/paper/2302.05950)

Discussion