DATA AUGMENTATION FOR NEURAL NETWORK OPTIMAL GENERALIZATION
| Autor: | Abdurashitova Muniskhon |
|---|---|
| Rok vydání: | 2023 |
| Předmět: | |
| DOI: | 10.5281/zenodo.7739564 |
| Popis: | To expand the size of a real dataset, data augmentation techniques artificially create various versions of the original dataset. Following their application, many techniques and methods have demonstrated an improvement in the precision of machine learning models. It serves as a regularizer during machine learning model training and aids in lowering overfitting. This article will cover the application of data augmentation techniques based on different noise types that shows improved neural network performance. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|