US 12,130,889 B2
Neural network training with homomorphic encryption
Nir Drucker, Haifa (IL); Ehud Aharoni, Kfar Saba (IL); Hayim Shaul, Kfar Saba (IL); Allon Adir, Kiryat Tivon (IL); and Lev Greenberg, Haifa (IL)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Mar. 21, 2022, as Appl. No. 17/655,566.
Prior Publication US 2023/0297649 A1, Sep. 21, 2023
Int. Cl. G06F 18/214 (2023.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); H04L 9/00 (2022.01)
CPC G06F 18/2148 (2023.01) [G06N 3/04 (2013.01); G06N 3/08 (2013.01); H04L 9/008 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method of training a neural network, the method comprising:
receiving, via an input to the neural network, a training dataset containing samples that are encrypted using homomorphic encryption;
determining a packing formation used by the homomorphic encryption used to pack the samples in the training dataset;
selecting a dropout technique during training of the neural network based on the packing formation; and
starting with a first packing formation from the training dataset, inputting the first packing formation in an iterative or recursive manner into the neural network using the selected dropout technique, with a next packing formation from the training dataset acting as an initial input that is applied to the neural network for a next iteration, until a stopping metric is produced by the neural network.