LOL: Learning To Optimize Loss Switching Under Label Noise
Ryuichiro Hataya, Hideki Nakayama · 2019
Deep convolutional neural networks excel in image recognition, but they are also known to be fragile to label corruption. To mitigate this problem, we propose to dynamically switch two loss functions, categorical cross entropy and mean absolute error, to exploit their complementary advantages. We employ the bilevel programming approach to simultaneously optimize base CNNs and the weights of two loss functions. Our proposed method only requires little modification in the optimization process of the original supervised problem and is applicable to a wide variety of networks under label corruption. Further, our approach achieves on-par results with other state-of-the-art noise-tolerant learning methods.