V-EfficientNets: Vector-Valued Efficiently Scaled Convolutional Neural Network Models

TL;DR

V-EfficientNets extend EfficientNet models to process vector-valued data, achieving high accuracy in medical image classification while significantly reducing parameters and outperforming existing models.

Contribution

The paper introduces vector-valued EfficientNets, a novel neural network architecture that processes multidimensional data coherently, enhancing efficiency and accuracy in specialized tasks.

Findings

Achieved 99.46% accuracy on ALL-IDB2 dataset

Significantly reduced model parameters compared to EfficientNet

Outperformed state-of-the-art models in medical image classification

Abstract

EfficientNet models are convolutional neural networks optimized for parameter allocation by jointly balancing network width, depth, and resolution. Renowned for their exceptional accuracy, these models have become a standard for image classification tasks across diverse computer vision benchmarks. While traditional neural networks learn correlations between feature channels during training, vector-valued neural networks inherently treat multidimensional data as coherent entities, taking for granted the inter-channel relationships. This paper introduces vector-valued EfficientNets (V-EfficientNets), a novel extension of EfficientNet designed to process arbitrary vector-valued data. The proposed models are evaluated on a medical image classification task, achieving an average accuracy of 99.46% on the ALL-IDB2 dataset for detecting acute lymphoblastic leukemia. V-EfficientNets demonstrate remarkable efficiency, significantly reducing parameters while outperforming state-of-the-art models, including the original EfficientNet. The source code is available at https://github.com/mevalle/v-nets.