A comprehensive theoretical framework for the optimization of neural networks classification performance with respect to weighted metrics
Open Access
- 25 April 2024
- journal article
- research article
- Published by Springer Nature in Optimization Letters
- Vol. 19 (1) , 169-192
Abstract
In many contexts, customized and weighted classification scores are designed in order to evaluate the goodness of the predictions carried out by neural networks. However, there exists a discrepancy between the maximization of such scores and the minimization of the loss function in the training phase. In this paper, we provide a complete theoretical setting that formalizes weighted classification metrics and then allows the construction of losses that drive the model to optimize these metrics of interest. After a detailed theoretical analysis, we show that our framework includes as particular instances well-established approaches such as classical cost-sensitive learning, weighted cross entropy loss functions and value-weighted skill scores.Keywords
All Related Versions
Funding Information
- Università degli Studi di Padova
This publication has 23 references indexed in Scilit:
- Prediction of severe thunderstorm events with ensemble deep learning and radar dataScientific Reports, 2022
- Probabilistic Prediction of Dst Storms One‐Day‐Ahead Using Full‐Disk SoHO ImagesSpace Weather, 2022
- Score-Oriented Loss (SOL) functionsPattern Recognition, 2022
- Implementation paradigm for supervised flare forecasting studies: A deep learning application with video dataAstronomy & Astrophysics, 2022
- Balanced Softmax Cross-Entropy for Incremental Learning2021
- A New Loss Function for CNN Classifier Based on Predefined Evenly-Distributed Class CentroidsIEEE Access, 2019
- Learning from Imbalanced Data Sets with Weighted Cross-Entropy FunctionNeural Processing Letters, 2019
- Cost-Sensitive Learning2018
- Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep LearningCell, 2018
- Decision‐making from probability forecasts based on forecast valueMeteorological Applications, 2002