# Auto-encoders: reconstruction versus compression

arXiv: https://arxiv.org/abs/1403.7752 · HTML: https://mlsanity.com/paper/1403.7752/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Mar 2014.

## Built on by (most cited first, 3 of 3)

- [Learning-Driven Lossy Image Compression; A Comprehensive Survey](https://mlsanity.com/paper/2201.09240/) (2022)
- [In-context learning and Occam's razor](https://mlsanity.com/paper/2410.14086/) (2024)
- [Reducing Redundancy in the Bottleneck Representation of the Autoencoders](https://mlsanity.com/paper/2202.04629/) (2022)

## Abstract (arXiv)

We discuss the similarities and differences between training an auto-encoder to minimize the reconstruction error, and training the same auto-encoder to compress the data via a generative model. Minimizing a codelength for the data using an auto-encoder is equivalent to minimizing the reconstruction error plus some correcting terms which have an interpretation as either a denoising or contractive property of the decoding function. These terms are related but not identical to those used in denoising or contractive auto-encoders [Vincent et al. 2010, Rifai et al. 2011]. In particular, the codelength viewpoint fully determines an optimal noise level for the denoising criterion.