Timezone: »
While deep neural networks achieve great performance on fitting the training distribution, the learned networks are prone to overfitting and are susceptible to adversarial attacks. In this regard, a number of mixup based augmentation methods have been recently proposed. However, these approaches mainly focus on creating previously unseen virtual examples and can sometimes provide misleading supervisory signal to the network. To this end, we propose Puzzle Mix, a mixup method for explicitly utilizing the saliency information and the underlying statistics of the natural examples. This leads to an interesting optimization problem alternating between the multi-label objective for optimal mixing mask and saliency discounted optimal transport objective. Our experiments show Puzzle Mix achieves the state of the art generalization and the adversarial robustness results compared to other mixup methods on CIFAR-100, Tiny-ImageNet, and ImageNet datasets, and the source code is available at https://github.com/snu-mllab/PuzzleMix.
Author Information
Jang-Hyun Kim (Seoul National University)
Wonho Choo (Seoul National University)
Hyun Oh Song (Seoul National University)
More from the Same Authors
-
2023 Poster: Efficient Latency-Aware CNN Depth Compression via Two-Stage Dynamic Programming »
Jinuk Kim · Yeonwoo Jeong · Deokjae Lee · Hyun Oh Song -
2022 Poster: Dataset Condensation via Efficient Synthetic-Data Parameterization »
Jang-Hyun Kim · Jinuk Kim · Seong Joon Oh · Sangdoo Yun · Hwanjun Song · Joonhyun Jeong · Jung-Woo Ha · Hyun Oh Song -
2022 Spotlight: Dataset Condensation via Efficient Synthetic-Data Parameterization »
Jang-Hyun Kim · Jinuk Kim · Seong Joon Oh · Sangdoo Yun · Hwanjun Song · Joonhyun Jeong · Jung-Woo Ha · Hyun Oh Song -
2022 Poster: Query-Efficient and Scalable Black-Box Adversarial Attacks on Discrete Sequential Data via Bayesian Optimization »
Deokjae Lee · Seungyong Moon · Junhyeok Lee · Hyun Oh Song -
2022 Spotlight: Query-Efficient and Scalable Black-Box Adversarial Attacks on Discrete Sequential Data via Bayesian Optimization »
Deokjae Lee · Seungyong Moon · Junhyeok Lee · Hyun Oh Song -
2019 Poster: Learning Discrete and Continuous Factors of Data via Alternating Disentanglement »
Yeonwoo Jeong · Hyun Oh Song -
2019 Oral: Learning Discrete and Continuous Factors of Data via Alternating Disentanglement »
Yeonwoo Jeong · Hyun Oh Song -
2019 Poster: Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimization »
Seungyong Moon · Gaon An · Hyun Oh Song -
2019 Oral: Parsimonious Black-Box Adversarial Attacks via Efficient Combinatorial Optimization »
Seungyong Moon · Gaon An · Hyun Oh Song -
2019 Poster: EMI: Exploration with Mutual Information »
Hyoungseok Kim · Jaekyeom Kim · Yeonwoo Jeong · Sergey Levine · Hyun Oh Song -
2019 Oral: EMI: Exploration with Mutual Information »
Hyoungseok Kim · Jaekyeom Kim · Yeonwoo Jeong · Sergey Levine · Hyun Oh Song -
2018 Poster: Efficient end-to-end learning for quantizable representations »
Yeonwoo Jeong · Hyun Oh Song -
2018 Oral: Efficient end-to-end learning for quantizable representations »
Yeonwoo Jeong · Hyun Oh Song