Maximum Entropy Auto-Encoding
This is an incremental improvement for auto-encoder models in machine learning, particularly for reconstruction tasks.
The paper tackled the problem of improving auto-encoder reconstruction by using optimal reconstruction based on a maximum entropy prior, resulting in up to a factor of two reduction in mean square reconstruction error.
In this paper, it is shown that an auto-encoder using optimal reconstruction significantly outperforms a conventional auto-encoder. Optimal reconstruction uses the conditional mean of the input given the features, under a maximum entropy prior distribution. The optimal reconstruction network, which is called deterministic projected belied network (D-PBN), resembles a standard reconstruction network, but with special non-linearities that mist be iteratively solved. The method, which can be seen as a generalization of maximum entropy image reconstruction, extends to multiple layers. In experiments, mean square reconstruction error reduced by up to a factor of two. The performance improvement diminishes for deeper networks, or for input data with unconstrained values (Gaussian assumption).