Bio

Education

• B.S. in Automation, Tsinghua University, 2011-2015
• M.S. in Automation, Tsinghua University, 2015-2018
• Ph.D in Automation, Tsinghua University, 2018-2021 (expected)

Experience

• Nov 2019 - Present: Research Intern
  • ByteDance AI Lab
  • Improve query efficiency for hard-label black-box attack using reinforcement learning
  • Supervisor: Yiwen Guo
• Sep 2018 - Jun 2019: Research Intern
  • Intel Labs China
  • Improve DNN’s adversarial robustness && more query efficient black-box attack
  • Supervisor: Jianguo Li and Yiwen Guo
  • Publication: Subspace Attack (NeurIPS 2019) and AMM (T-PAMI 2019)
• Mar 2018 - Jun 2018: BlockChain Developer
  • Create an ETH game named AlchemyGod with my classmates
  • Create a trading bot on main-stream crypto currencies (BTC, LTC and ETH)
• Aug 2017 - May 2018: Research Assistant
  • Tsinghua University && Intel Labs China
  • Defend against adversarial attacks by forcing attackers to increase perturbation norm
  • Supervisor: Yiwen Guo
  • Publication: DeepDefense (NeurIPS 2018)
• 2015 - 2017: Research Assistant
  • Department of Automation, Tsinghua University
  • Traffic sign detection && weakly supervised object detection && zero shot learning
  • Supervisor: Changshui Zhang
• Summer 2014: Research Intern
  • Carnegie Mellon University
  • Robot navigation
  • Supervisor: Hernan Badino and Daniel Huber
• 2013 - 2014: Research Assistant
  • Tsinghua University
  • Rule-based high-level decision making, RoboCup Tsinghua Hephaetus Team
  • Supervisor: Mingguo Zhao

Publications

Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples

Yan, Ziang, Yiwen Guo, Jian Liang, and Changshui Zhang. "Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples." ICLR 2021.

Subspace Attack: Exploiting Promising Subspaces for Query-Efficient Black-box Attacks

Yan, Ziang*, Yiwen Guo*, and Changshui Zhang. "Subspace Attack: Exploiting Promising Subspaces for Query-Efficient Black-box Attacks." NeurIPS 2019.

Adversarial Margin Maximization Networks

Yan, Ziang, Yiwen Guo, and Changshui Zhang. "Adversarial Margin Maximization Networks." PAMI 2019.

Deep Defense: Training DNNs with Improved Adversarial Robustness

Yan, Ziang*, Yiwen Guo*, and Changshui Zhang. "DeepDefense: Training Deep Neural Networks with Improved Robustness." NeurIPS 2018.