Showing 1–1 of 1 results for author: Yasinnik, B
-
Imbalanced Classification via Explicit Gradient Learning From Augmented Data
Authors:
Bronislav Yasinnik,
Moshe Salhov,
Ofir Lindenbaum,
Amir Averbuch
Abstract:
Learning from imbalanced data is one of the most significant challenges in real-world classification tasks. In such cases, neural networks performance is substantially impaired due to preference towards the majority class. Existing approaches attempt to eliminate the bias through data re-sampling or re-weighting the loss in the learning process. Still, these methods tend to overfit the minority sa…
▽ More
Learning from imbalanced data is one of the most significant challenges in real-world classification tasks. In such cases, neural networks performance is substantially impaired due to preference towards the majority class. Existing approaches attempt to eliminate the bias through data re-sampling or re-weighting the loss in the learning process. Still, these methods tend to overfit the minority samples and perform poorly when the structure of the minority class is highly irregular. Here, we propose a novel deep meta-learning technique to augment a given imbalanced dataset with new minority instances. These additional data are incorporated in the classifier's deep-learning process, and their contributions are learned explicitly. The advantage of the proposed method is demonstrated on synthetic and real-world datasets with various imbalance ratios.
△ Less
Submitted 28 October, 2022; v1 submitted 21 February, 2022;
originally announced February 2022.
