Deep Transform: Error Correction via Probabilistic Re-Synthesis
This addresses the challenge of defining and correcting errors without supervision, which is incremental as it builds on existing deep learning methods for data synthesis.
The paper tackles the problem of unsupervised error correction in data by training a deep neural network to re-synthesize inputs, using a deep transform to reject errors outside the feature space, and demonstrates recovery from extreme degradation.
Errors in data are usually unwelcome and so some means to correct them is useful. However, it is difficult to define, detect or correct errors in an unsupervised way. Here, we train a deep neural network to re-synthesize its inputs at its output layer for a given class of data. We then exploit the fact that this abstract transformation, which we call a deep transform (DT), inherently rejects information (errors) existing outside of the abstract feature space. Using the DT to perform probabilistic re-synthesis, we demonstrate the recovery of data that has been subject to extreme degradation.