search

csong27 / membership-inference

Code for Membership Inference Attack against Machine Learning Models (in Oakland 2017)
176 stars 63 forks source link

How to optimize the parameter #2

Open knightcao opened 5 years ago

knightcao commented 5 years ago

Hello, Song. I try your code via UCI adult's salary data set and cifar10. I get a result similar with yours for adult's salary data set, however for cifar10, i get the a very low accuracy. I guess maybe some parameter is need optimized, such as batch_size or epochs.
So, I wonder if you willing to publish your parameters ? And another question is that for images, I should replace the Target's and Shadow's models from NN to CNN or other? Thank you, I really appreciate your work.

csong27 commented 5 years ago

Hi, On cifar10 you will need a bigger neural nets than the one in the demo code. CNN is a good idea :)

knightcao commented 5 years ago

Thank you, happy voteday.

knightcao commented 5 years ago

I found under fitting will take place with default parameter in most dataset. I dont know whether properly or not? Thank you.

csong27 commented 5 years ago

You will have to fit each dataset with its own best hyper-parameters.

knightcao commented 5 years ago

Thank you, have a good weekend.