US 12,271,805 B1
Data obfuscation using encoder-multi-decoder architecture
Julian Collado Umana, Irvine, CA (US); and Andrew Davis, Portland, OR (US)
Assigned to HiddenLayer, Inc., Austin, TX (US)
Filed by HiddenLayer, Inc., Austin, TX (US)
Filed on Sep. 20, 2024, as Appl. No. 18/892,215.
Int. Cl. G06N 3/0455 (2023.01); G06N 3/084 (2023.01); H04L 9/08 (2006.01)
CPC G06N 3/0455 (2023.01) [G06N 3/084 (2013.01); H04L 9/0819 (2013.01)] 27 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving, by an encoder, first data encapsulating second data in a hidden compartment and a decoder identifier corresponding to either of a first decoder or a second decoder;
generating, by the encoder, an embedding corresponding to the first data;
decoding, by the first decoder, the embedding to result in a representation of the first data when the decoder identifier corresponds to the first decoder;
decoding, by the second decoder, the embedding to result in a representation of the second data when the decoder identifier corresponds to the second decoder; and
providing the decoded embedding to a consuming application or process;
wherein the first decoder and second decoder each comprise a respective neural network.