Using Autoencoder for compressions?

General Discussion
#1

Is there a way to introduce sparsity Fourier/Cosine/others constraint on the weights of an autoencoder to achieve compressions? I want to use the encoder part to replicate a better basis with the STL-10 dataset. Autoencoders | Kaggle The features learned should be more optimized than the compression algorithms like JPEG.

#3

I suggest to take an overview at:
https://arxiv.org/abs/2002.03711

Then you can take a look at the CLIC 21 tasks/leaderboard:

http://compression.cc/

See also our repository at:

2 Likes
#5

Thanks, will look into it. I am curious if I can replace L1 regularization with autoencoder?

#6

The first link two links are specifically related to the image (or video) compression tasks.

The TF compression repository instead it is related to multiple data compression use cases.

I am curious if I can replace L1 regularization with autoencoder?

Generally you can use also KL-divergence as sparsity penalty.

2 Likes