One pixel attack for fooling deep neural networks

From statwiki
Revision as of 20:52, 25 March 2018 by Mhlu (talk | contribs) (Methodology)
Jump to: navigation, search

Presented by

1. Ziheng Chu

2. Minghao Lu

3. Qi Mai

4. Qici Tan

Introduction

Methodology

We first formalize the generation of adversarial images as an constrained optimization problem. Let [math]\bold{x}\lt math\gt be the vectorized form of an image. Let \lt math\gt f_t(\bold{x})[/math]

Evaluation and Results

Measurement Metrics and Datasets

Results

Discussion and Future Work