[2019-01-28] Prof. Xu-Ying Liu, Southeast University, "Imbalanced Augmented Class Learning with Unlabeled Data by Label Confidence Propagation"

Poster:Post date:2019-01-21
Title: Imbalanced Augmented Class Learning with Unlabeled Data by Label Confidence Propagation
Date: 2019-01-28 11:00am-12:00pm
Location: R105, CSIE
Speaker: Prof. Xu-Ying Liu, Southeast University
Hosted by: Prof. Cheng-Fu Chou


As a practical problem in open and dynamic environments, class-incremental learning has attracted much attention from many fields. Learning with augmented class (LAC) problem formulates one of the core difficulties of class-incremental learning: instances of augmented class need to be predicted with the restriction that only examples from seen classes are observed in training phase. LACU framework advances the study of LAC problem by exploiting unlabeled data, while it does not take into account an important practical problem widely-existing in real-world applications of LAC – imbalanced class distributions among seen classes, which will further increase the learning difficulties of LAC problem. We propose a novel approach Label Confidence Propagation (LCP) to tackle the problem of imbalanced augmented class learning with unlabeled data. LCP enlarges the labeled training data set by estimating class labels for unlabeled data, to meet the challenge of lacking supervision information of augmented classes via identifying some of their instances, and to alleviate the damage of class-imbalance via identifying more instances for each seen class. LCP firstly initializes label confidence, i.e., the posterior probability distributions of all classes (including augmented classes) for unlabeled data, then iteratively propagates label confidence to identify a valid label for each unlabeled instance to enlarge the labeled training data set. Finally, LCP predicts for unseen instances by linear neighborhood reconstruction to be robust to potential noise. Results on abundant experiments show that LCP is significantly superior to many state-of-the-art methods, and robust to high imbalance ratio and high open level. LCP can sufficiently unleash its strength especially when there are abundant unlabeled data available.

Xu-Ying Liu is an assistant professor at the School of Computer Science and Engineering of Southeast University, P. R. China. She received her Ph.D. degree in computer science from Nanjing University, P. R. China in 2010. Her research interests mainly include machine learning and data mining, especially cost-sensitive learning, class-imbalance learning and open category classification. She published 28 papers with more than 2400 citations, including papers in leading international journals or conferences. Her research work on class-imbalance learning has attracted wide attention in related fields. She is a member of CCF (China Computer Federation) Artificial Intelligence & Pattern Recognition Committee.
Last modification time:2019-01-21 AM 10:47

cron web_use_log