Deep Representation Debiasing via Mutual Information Minimization and Maximization (Student Abstract)

Deep representation learning has succeeded in several fields. However, pre-trained deep representations are usually biased and make downstream models sensitive to different attributes. In this work, we propose a post-processing unsupervised deep representation debiasing algorithm, DeepMin- Max, which can obtain unbiased representations directly from pre-trained representations without re-training or fine-tuning the entire model. The experimental results on synthetic and real-world datasets indicate that DeepMinMax outperforms the existing state-of-the-art algorithms on downstream tasks.

Authors

• Ruijiang Han
• Wei Wang
• Yuxi Long
• Jiajie Peng AI for Science Interdisciplinary Research Center, School of Computer Science, Northwestern Polytechnical University School of Computer Science, Research and Development Institute of Northwestern Polytechnical University in Shenzhen Key Laboratory of Big Data Storage and Management, Northwestern Polytechnical University, Ministry of Industry and Information Technology

Keywords

No keywords are indexed for this paper.