Baseline Pruning-Based Approach to Trojan Detection in Neural Networks

01/22/2021
by   Peter Bajcsy, et al.
1

This paper addresses the problem of detecting trojans in neural networks (NNs) by analyzing systematically pruned NN models. Our pruning-based approach consists of three main steps. First, detect any deviations from the reference look-up tables of model file sizes and model graphs. Next, measure the accuracy of a set of systematically pruned NN models following multiple pruning schemas. Finally, classify a NN model as clean or poisoned by applying a mapping between accuracy measurements and NN model labels. This work outlines a theoretical and experimental framework for finding the optimal mapping over a large search space of pruning parameters. Based on our experiments using Round 1 and Round 2 TrojAI Challenge datasets, the approach achieves average classification accuracy of 69.73 less than 60 s per model. For both datasets random guessing would produce 50 classification accuracy. Reference model graphs and source code are available from GitHub.

READ FULL TEXT

page 1

page 2

page 3

page 4

page 5

page 6

research
02/10/2021

Pruning of Convolutional Neural Networks Using Ising Energy Model

Pruning is one of the major methods to compress deep neural networks. In...
research
05/13/2020

Artificial Neural Network Pruning to Extract Knowledge

Artificial Neural Networks (NN) are widely used for solving complex prob...
research
11/17/2020

Dynamic Hard Pruning of Neural Networks at the Edge of the Internet

Neural Networks (NN), although successfully applied to several Artificia...
research
07/06/2020

EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruning

Finding out the computational redundant part of a trained Deep Neural Ne...
research
07/23/2023

RANSAC-NN: Unsupervised Image Outlier Detection using RANSAC

Image outlier detection (OD) is crucial for ensuring the quality and acc...
research
12/19/2021

On Causal Inference for Data-free Structured Pruning

Neural networks (NNs) are making a large impact both on research and ind...
research
05/23/2017

An effective algorithm for hyperparameter optimization of neural networks

A major challenge in designing neural network (NN) systems is to determi...

Please sign up or login with your details

Forgot password? Click here to reset