GERIS: A Game-Theoretic Framework for Filtering Instance-Dependent Label Noise in License Plate Data Augmentation
Experimental results demonstrate that GERIS outperforms existing instance selection methods in terms of classification accuracy and robustness, and offers a transparent, theoretically grounded mechanism for data filtering.
S. Shani, Rouhollah Ahmadian, Amin Rahmani et al.
· 0 citations