Hyeonjin Park
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Hyeonjin Park has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Donggeun Ko
AiM Future, Seoul, Republic of Korea
1 shared publication - Dongjun Lee
Maum AI, Gyeonggi-do, Republic of Korea
1 shared publication - Jaekwang Kim
Sungkyunkwan University
1 shared publication - Kyoungrae Noh
Sungkyunkwan University, Suwon, Republic of Korea
1 shared publication - Namjun Park
Department of Computer Science and Engineering, Sungkyunkwan University, Suwon 16414, Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 4028-4032
AmpliBias: Mitigating Dataset Bias through Bias Amplification in Few-shot Learning for Generative Models
Deep learning models exhibit a dependency on peripheral attributes of input data, such as shapes and colors, leading the models to become biased towards these certain attributes that result in subsequent degradation of performance. In this paper, we alleviate this problem by presenting~\sysname, a novel framework that tackles dataset bias by leveraging generative models to amplify bias and facilitate the learning of debiased representations of the classifier. Our method involves three major steps. We initially train a biased classifier, denoted as f_b, on a biased dataset and extract the top-K biased-conflict samples. Next, we train a generator solely on a bias-conflict dataset comprised of these top-K samples, aiming to learn the distribution of bias-conflict samples. Finally, we re-train the classifier on the newly constructed debiased dataset, which combines the original and amplified data. This allows the biased classifier to competently learn debiased representation. Extensive experiments validate that our proposed method effectively debiases the biased classifier.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 4028-4032
Open AccessAmpliBias: Mitigating Dataset Bias through Bias Amplification in Few-shot Learning for Generative Models
Deep learning models exhibit a dependency on peripheral attributes of input data, such as shapes and colors, leading the models to become biased towards these certain attributes that result in subsequent degradation of performance. In this paper, we alleviate this problem by presenting~\sysname, a novel framework that tackles dataset bias by leveraging generative models to amplify bias and facilitate the learning of debiased representations of the classifier. Our method involves three major steps. We initially train a biased classifier, denoted as f_b, on a biased dataset and extract the top-K biased-conflict samples. Next, we train a generator solely on a bias-conflict dataset comprised of these top-K samples, aiming to learn the distribution of bias-conflict samples. Finally, we re-train the classifier on the newly constructed debiased dataset, which combines the original and amplified data. This allows the biased classifier to competently learn debiased representation. Extensive experiments validate that our proposed method effectively debiases the biased classifier.