Research Article

Robust Fully Distributed Minibatch Gradient Descent with Privacy Preservation

Table 1


Parameter setupsTime consumption (seconds)Results
ā€‰Number of features ()Max tree size ()Bits per feature ()Key size ()Blocks Message size to parent Encrypt/decrypt a blockSend plain- text modelEncrypt sharesOne aggregation roundOverall time of mini-batchProb. of good tree

4419101024181920.0410.1030.1230.1431.8470.999
20481163840.3000.1030.9000.4044.4510.997
6671410242163840.0410.1030.2460.1862.8500.997
20481163840.3000.1030.9000.4045.4660.996
419101024998110080.0410.42012.1774.36245.6490.969
2048508192000.3000.42045.00015.305155.0740.904
66714102413711223040.0410.42016.8515.99874.6090.951
20486911304960.3000.42062.10021.083255.6240.850