Towards better data augmentation using Wasserstein distance in variational auto-encoder
VAE, or variational auto-encoder, compresses data into latent attributes, and generates new data of different varieties. VAE based on KL divergence has been considered as an effective technique for data augmentation. In this paper, we propose the use of Wasserstein distance as a measure of distribut...
محفوظ في:
| المؤلفون الرئيسيون: | , |
|---|---|
| التنسيق: | text |
| اللغة: | English |
| منشور في: |
Institutional Knowledge at Singapore Management University
2021
|
| الموضوعات: | |
| الوصول للمادة أونلاين: | https://ink.library.smu.edu.sg/lkcsb_research/7046 https://ink.library.smu.edu.sg/context/lkcsb_research/article/8045/viewcontent/2109.14795.pdf |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| المؤسسة: | Singapore Management University |
| اللغة: | English |
| الملخص: | VAE, or variational auto-encoder, compresses data into latent attributes, and generates new data of different varieties. VAE based on KL divergence has been considered as an effective technique for data augmentation. In this paper, we propose the use of Wasserstein distance as a measure of distributional similarity for the latent attributes, and show its superior theoretical lower bound (ELBO) compared with that of KL divergence under mild conditions. Using multiple experiments, we demonstrate that the new loss function exhibits better convergence property and generates artificial images that could better aid the image classification tasks. |
|---|