Data cleaning for A.s.a. H.

 Many of our experiments produce noisy inputs. We sometimes use an auto-encoder neural network for data cleaning as a preprocessor. Whenever possible this is pretrained using clean data sets.