Cultivating Robustness for Deep Learning

Abstract

In this project we report our research activities on exploring the vulnerability and improving the robustness of deep learning against adversarial attacks. To be specific, we systematically analyze and investigate the efficient attack and defense mechanism of DNN models across different domains for different types of models with different constraints . In addition, the efficient interpretation and verification are also studied and developed, thereby providing useful perspective to detect and check the correctness of the input and DNN models.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 19, 2022
Accession Number
AD1156543

Entities

People

  • Bo Yuan
  • Myung Lee
  • Xue Lin

Organizations

  • Research Foundation of The City University of New York

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Air Force
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Automated Speech Recognition
  • Birds
  • Computational Science
  • Computer Languages
  • Computer Programming
  • Computer Vision
  • Computers
  • Data Mining
  • Detection
  • Information Systems
  • Machine Learning
  • Neural Networks
  • Pattern Recognition
  • Social Media

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Cybersecurity.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks