TY  - CONF
AU  - Bhalaji Nagarajan
AU  - Ricardo Marques
AU  - Marcos Mejia
AU  - Petia Radeva
A2  - VISAPP
PY  - 2022//
TI  - Class-conditional Importance Weighting for Deep Learning with Noisy Labels
BT  - 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
SP  - 679
EP  - 686
VL  - 5
KW  - Noisy Labeling
KW  - Loss Correction
KW  - Class-conditional Importance Weighting
KW  - Learning with Noisy Labels
N2  - Large-scale accurate labels are very important to the Deep Neural Networks to train them and assure high performance. However, it is very expensive to create a clean dataset since usually it relies on human interaction. To this purpose, the labelling process is made cheap with a trade-off of having noisy labels. Learning with Noisy Labels is an active area of research being at the same time very challenging. The recent advances in Self-supervised learning and robust loss functions have helped in advancing noisy label research. In this paper, we propose a loss correction method that relies on dynamic weights computed based on the model training. We extend the existing Contrast to Divide algorithm coupled with DivideMix using a new class-conditional weighted scheme. We validate the method using the standard noise experiments and achieved encouraging results.
UR  - https://www.scitepress.org/Link.aspx?doi=10.5220/0010996400003124
UR  - http://dx.doi.org/10.5220/0010996400003124
N1  - MILAB; no menciona
ID  - Bhalaji Nagarajan2022
ER  -