On Adversarial Mixup Resynthesis
By Christopher Beckham, Sina Honari, Alex Lamb, Vikas Verma, Farnoosh Ghadiri, R. Devon Hjelm, Yoshua Bengio, C. Pal. 2019
75 citations recorded by the source corpus.
Subfields: Optimization & Training
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Optimization & Training
- Hieu Pham
- Minh-Thang Luong
- Honglak Lee
- Kihyuk Sohn
- Eduard H. Hovy
- Jinwoo Shin
- Jitendra Malik
- Zhun Deng
Citations in the corpus
- A Survey of Data Augmentation Approaches for NLP (2021, background): “, 2020), more general mixing schemes (Guo, 2020), and adding adversaries (Beckham et al., 2019).”
- How Does Mixup Help With Robustness and Generalization? (2020, methodology): “…future direction is to extend our analysis to other Mixup variants, for example, Puzzle Mix (Kim et al., 2020) and Adversarial Mixup Resynthesis (Beckham et al., 2019), and investigate if the generalization performance and adversarial robustness can be further improved by these newly developed…”
- Multi-SpectroGAN: High-Diversity and High-Fidelity Spectrogram Generation with Adversarial Style Combination for Speech Synthesis (2020, methodology): “The method for learning combined latent representation of autoencoder was proposed (Beckham et al. 2019).”
- Inclusive GAN: Improving Data and Minority Coverage in Generative Models (2020, methodology): “ent the matching latent vectors with interpolation. The benets of interpolation-based data augmentation are validated by previous works in both discriminative modeling [50,4] and generative modeling [3,48]. Second, to make the metric spaces of the two losses more comparable, we measure the reconstru”
- Towards Domain-Agnostic Contrastive Learning (2020, background): “For the hyperparameter ρ of Binary-Mixup (Section 3.1), we search the value from the set [0.1, 0.3, 0.5].”
- i-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning (2021, background): “Other variations have also been investigated by interpolating in the feature space (Verma et al., 2019) or leveraging domain knowledge (Yun et al., 2019).”