HOW SAMPLING AFFECTS TRAINING: AN EFFECTIVE SAMPLING THEORY STUDY FOR LONG-TAILED IMAGE CLASSIFICATION

Gong Zhang, Yongqiang Gao, Liu Haijing

Abstract: The long-tailed image classification problem has been very challenging for a longtime. Suffered from the unbalanced distribution of categories, many deep vision classification methods perform well in the head classes while poor in the tail ones. This paper proposes an effective sampling theory, attempting to provide a theoretical explanation for the decoupling representation and classifier for long-tailed image classification. To apply the above sampling theory in practice, a general jitter sampling strategy is proposed. Experiments show that variety of long-tailed distribution algorithms exhibit better performance based on the effective sampling theory. The code will be released soon later.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning