Training Autoencoders in Sparse Domain

Biswarup Bhattacharya, Arna Ghosh, Somnath Basu Roy Chowdhury

AAAI 2018 pp. 8049-8050

doi:10.1609/AAAI.V32I1.12155 /aaai/2018/bhattacharya2018aaai-training/

Abstract

Autoencoders (AE) are essential in learning representation of large data (like images) for dimensionality reduction. Images are converted to sparse domain using transforms like Fast Fourier Transform (FFT) or Discrete Cosine Transform (DCT) where information that requires encoding is minimal. By optimally selecting the feature-rich frequencies, we are able to learn the latent vectors more robustly. We successfully show enhanced performance of autoencoders in sparse domain for images.

PDF AAAI Semantic Scholar

Cite

Text

Bhattacharya et al. "Training Autoencoders in Sparse Domain." AAAI Conference on Artificial Intelligence, 2018. doi:10.1609/AAAI.V32I1.12155

Markdown

[Bhattacharya et al. "Training Autoencoders in Sparse Domain." AAAI Conference on Artificial Intelligence, 2018.](https://mlanthology.org/aaai/2018/bhattacharya2018aaai-training/) doi:10.1609/AAAI.V32I1.12155

BibTeX

@inproceedings{bhattacharya2018aaai-training,
  title     = {{Training Autoencoders in Sparse Domain}},
  author    = {Bhattacharya, Biswarup and Ghosh, Arna and Chowdhury, Somnath Basu Roy},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {8049-8050},
  doi       = {10.1609/AAAI.V32I1.12155},
  url       = {https://mlanthology.org/aaai/2018/bhattacharya2018aaai-training/}
}