Timezone: »

 
Poster
Instance-dependent Label-noise Learning under a Structural Causal Model
Yu Yao · Tongliang Liu · Mingming Gong · Bo Han · Gang Niu · Kun Zhang

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @
in Poster Session 1 »
Label noise generally degenerates the performance of deep learning algorithms because deep neural networks easily overfit label errors. Let $X$ and $Y$ denote the instance and clean label, respectively. When $Y$ is a cause of $X$, according to which many datasets have been constructed, e.g., \textit{SVHN} and \textit{CIFAR}, the distributions of $P(X)$ and $P(Y|X)$ are generally entangled. This means that the unsupervised instances are helpful to learn the classifier and thus reduce the side effect of label noise. However, it remains elusive on how to exploit the causal information to handle the label-noise problem. We propose to model and make use of the causal process in order to correct the label-noise effect.Empirically, the proposed method outperforms all state-of-the-art methods on both synthetic and real-world label-noise datasets.
Keywords: Deep Learning · Causality

Author Information

Yu Yao (University of Sydney)
Tongliang Liu (The University of Sydney)
Mingming Gong (University of Melbourne)
Bo Han (HKBU / RIKEN)
Gang Niu (RIKEN)
Kun Zhang (CMU)

More from the Same Authors