Details

Autor(en) / Beteiligte

• Lingxi Xie
• Jingdong Wang
• Zhen Wei
• Meng Wang
• Qi Tian

Titel

DisturbLabel: Regularizing CNN on the Loss Layer

Ist Teil von

• 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, p.4753-4762

Ort / Verlag

IEEE

Erscheinungsjahr

2016

Quelle

IEEE Electronic Library (IEL)

Beschreibungen/Notizen

• During a long period of time we are combating overfitting in the CNN training process with model regularization, including weight decay, model averaging, data augmentation, etc. In this paper, we present DisturbLabel, an extremely simple algorithm which randomly replaces a part of labels as incorrect values in each iteration. Although it seems weird to intentionally generate incorrect training labels, we show that DisturbLabel prevents the network training from over-fitting by implicitly averaging over exponentially many networks which are trained with different label sets. To the best of our knowledge, DisturbLabel serves as the first work which adds noises on the loss layer. Meanwhile, DisturbLabel cooperates well with Dropout to provide complementary regularization functions. Experiments demonstrate competitive recognition results on several popular image recognition datasets.