From 755de4794864265a2115ad1de890dd3b7ba87d36 Mon Sep 17 00:00:00 2001 From: chandramouli1 Date: Tue, 8 Jun 2021 17:58:14 +0530 Subject: [PATCH] Solution for Learned Data Augmentation. --- _requests_for_research/learned-data-augmentation.html | 4 ++++ 1 file changed, 4 insertions(+) create mode 100644 _requests_for_research/learned-data-augmentation.html diff --git a/_requests_for_research/learned-data-augmentation.html b/_requests_for_research/learned-data-augmentation.html new file mode 100644 index 0000000..a57ce9e --- /dev/null +++ b/_requests_for_research/learned-data-augmentation.html @@ -0,0 +1,4 @@ +

⭐⭐⭐ Learned Data Augmentation. You could use a learned VAE of data, to perform “learned data augmentation”. One would first train a VAE on input data, then each training point would be transformed by encoding to a latent space, then applying a simple (e.g. Gaussian) perturbation in latent space, then decoding back to observed space. Could we use such an approach to obtain improved generalization? A potential benefit of such data augmentation is that it could include many nonlinear transformations like viewpoint changes and changes in scene lightning. Can we approximate the set of transformations to which the label is invariant? Check out the existing work on this topic if you want a place to get started.

+ +

Solutions

+

Preliminary paper describing learned data augmentation can be found here , with the implementation avaiable at this repo . Current results improve the generalization of the VAE when tested on the CIFAR-10 dataset.

\ No newline at end of file