We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Does Adversarial Oversampling Help us?

Abstract: Traditional oversampling methods are generally employed to handle class imbalance in datasets. This oversampling approach is independent of the classifier; thus, it does not offer an end-to-end solution. To overcome this, we propose a three-player adversarial game-based end-to-end method, where a domain-constraints mixture of generators, a discriminator, and a multi-class classifier are used. Rather than adversarial minority oversampling, we propose an adversarial oversampling (AO) and a data-space oversampling (DO) approach. In AO, the generator updates by fooling both the classifier and discriminator, however, in DO, it updates by favoring the classifier and fooling the discriminator. While updating the classifier, it considers both the real and synthetically generated samples in AO. But, in DO, it favors the real samples and fools the subset class-specific generated samples. To mitigate the biases of a classifier towards the majority class, minority samples are over-sampled at a fractional rate. Such implementation is shown to provide more robust classification boundaries. The effectiveness of our proposed method has been validated with high-dimensional, highly imbalanced and large-scale multi-class tabular datasets. The results as measured by average class specific accuracy (ACSA) clearly indicate that the proposed method provides better classification accuracy (improvement in the range of 0.7% to 49.27%) as compared to the baseline classifier.
Comments: Paper accepted in CIKM 2021
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2108.10697 [cs.LG]
  (or arXiv:2108.10697v1 [cs.LG] for this version)

Submission history

From: Tanmoy Dam [view email]
[v1] Fri, 20 Aug 2021 05:43:17 GMT (108kb,D)

Link back to: arXiv, form interface, contact.