Timezone: »
Label space dimension reduction (LSDR) is an efficient and effective paradigm for multi-label classification with many classes. Existing approaches to LSDR, such as compressive sensing and principal label space transformation, exploit only the label part of the dataset, but not the feature part. In this paper, we propose a novel approach to LSDR that considers both the label and the feature parts. The approach, called conditional principal label space transformation, is based on minimizing an upper bound of the popular Hamming loss. The minimization step of the approach can be carried out efficiently by a simple use of singular value decomposition. In addition, the approach can be extended to a kernelized version that allows the use of sophisticated feature combinations to assist LSDR. The experimental results verify that the proposed approach is more effective than existing ones to LSDR across many real-world datasets.
Author Information
Yao-Nan Chen (National Taiwan University)
Hsuan-Tien Lin (National Taiwan University)
More from the Same Authors
-
2021 : Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration »
Si-An Chen · Chun-Liang Li · Hsuan-Tien Lin -
2021 : On the Role of Pre-training for Meta Few-Shot Learning »
Chia-You Chen · Hsuan-Tien Lin · Masashi Sugiyama · Gang Niu -
2021 Meetup: Taipei, Taiwan »
Hsuan-Tien Lin -
2021 : Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration »
Si-An Chen · Chun-Liang Li · Hsuan-Tien Lin -
2022 : Improving Conditional Score-Based Generation with Calibrated Classification and Joint Training »
Paul K. Huang · Si-An Chen · Hsuan-Tien Lin -
2021 Poster: A Unified View of cGANs with and without Classifiers »
Si-An Chen · Chun-Liang Li · Hsuan-Tien Lin -
2018 Poster: REFUEL: Exploring Sparse Features in Deep Reinforcement Learning for Fast Disease Diagnosis »
Yu-Shao Peng · Kai-Fu Tang · Hsuan-Tien Lin · Edward Chang -
2006 Poster: Ordinal Regression by Extended Binary Classification »
Ling Li · Hsuan-Tien Lin -
2006 Spotlight: Ordinal Regression by Extended Binary Classification »
Ling Li · Hsuan-Tien Lin