ALEXANDRIA, Va., Nov. 11 -- United States Patent no. 12,468,780, issued on Nov. 11, was assigned to Oracle International Corp. (Redwood Shores, Calif.).
"Balancing feature distributions using an importance factor" was invented by Matteo Casserini (Zurich), Saeid Allahdadian (Vancouver, Canada), Felix Schmidt (Baden-Daettwil, Switzerland) and Andrew Brownsword (Vancouver, Canada).
According to the abstract* released by the U.S. Patent & Trademark Office: "Herein are machine learning techniques that adjust reconstruction loss of a reconstructive model such as an autoencoder based on importances of values of features. In an embodiment and before, during, or after training, the reconstructive model that more or less accurately reconstructs it...