Symmetrization of Loss Functions for Robust Training of Neural Networks in the Presence of Noisy Labels
Researchers have developed a new method for training neural networks that is more robust to errors in labeled data. This approach, called symmetrization of loss functions, theoretically guarantees better performance when dealing with noisy labels. The study introduces specific multi-class loss functions, including SGCE and alpha-MAE, which interpolate between existing methods and offer control over smoothness, showing competitive results on benchmarks. AI
IMPACT Introduces a novel technique to improve the reliability of machine learning models trained on imperfect datasets.