Skip to content

Explore impact of network architecture on attack efficacy #2

Description

@TTitcombe

Vary:

  • number of hidden layers
  • size of hidden layers
  • size of intermediate tensor (sent between model parts)
  • convolution v fully connected layers

From earlier work, should be that more layers -> attack power decreases; FC layers reduce attack power more than conv layers; smaller intermediate tensor -> attack power decreases

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    Projects

    No projects

      Milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions