Timezone: »

Deep Insights into Noisy Pseudo Labeling on Graph Data
Botao WANG · Jia Li · Yang Liu · Jiashun Cheng · Yu Rong · Wenjia Wang · Fugee Tsung

Wed Dec 13 03:00 PM -- 05:00 PM (PST) @ Great Hall & Hall B1+B2 #813

Pseudo labeling (PL) is a wide-applied strategy to enlarge the labeled dataset by self-annotating the potential samples during the training process. Several works have shown that it can improve the graph learning model performance in general. However, we notice that the incorrect labels can be fatal to the graph training process. Inappropriate PL may result in the performance degrading, especially on graph data where the noise can propagate. Surprisingly, the corresponding error is seldom theoretically analyzed in the literature. In this paper, we aim to give deep insights of PL on graph learning models. We first present the error analysis of PL strategy by showing that the error is bounded by the confidence of PL threshold and consistency of multi-view prediction. Then, we theoretically illustrate the effect of PL on convergence property. Based on the analysis, we propose a cautious pseudo labeling methodology in which we pseudo label the samples with highest confidence and multi-view consistency. Finally, extensive experiments demonstrate that the proposed strategy improves graph learning process and outperforms other PL strategies on link prediction and node classification tasks.

Author Information

Botao WANG (Hong Kong University of Science and Technology)

He received a Bachelor's and a Master's degree in Electrical Engineering from Xi'an Jiaotong University. He is now a PhD candidate of Data Science and Analytics in Hong Kong University of Science and Technology, supervised by Prof. Fugee Tsung and Prof. Wenjia Wang. His current research interests include industrial big data, anomaly detection and graph learning.

Jia Li
Yang Liu (HKUST)
Jiashun Cheng (Hong Kong University of Science and Technology)
Yu Rong (Shenzhen Tencent Computer System Co., Ltd.)
Wenjia Wang (SAMSI)
Fugee Tsung (Hong Kong University of Science and Technology)

More from the Same Authors