Jiaru Zhang

Jiaru Zhang

Ph.D Candidate at Shanghai Jiao Tong University

Shanghai Jiao Tong University

Biography

I am a 25-year-old Ph.D. Candidate of Computer Science at Shanghai Jiao Tong University, supervised by Prof. Haibing Guan, and co-supervised by Prof. Yang Hua, Prof. Tao Song, and Prof. Ruhui Ma. My research interests include Bayesian Neural Networks, Adversarial Attack and Defense and Causal Discovery.

I am looking for a postdoc position now.

Interests
  • Bayesian Neural Networks
  • Adversarial Attack and Defense
  • Causal Discovery
  • Diffusion Models
Education
  • PhD in Computer Science and Technology, 2019.9 - 2024.6 (expected)

    Shanghai Jiao Tong University

  • BSc in Computer Science and Technology (IEEE Honor Class), 2015.9 - 2019.6

    Shanghai Jiao Tong University

  • Electrical Engineering International Intensive program, 2017.6 - 2017.8

    University of Washington

Publication List

(2024). Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks. In Submission.

PDF

(2024). CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion. In CVPR.

(2024). SPOT: Harnessing Differentiable Causal Discovery in the Presence of Latent Confounders with Skeleton Posterior. In SIGKDD.

(2023). THOR: Generic Energy Estimation for On-device Training. In Submission.

Internship Experience

 
 
 
 
 
Microsoft Research Asia
Research Intern
Microsoft Research Asia
November 2022 – August 2023 Beijing

Mentor: Justin Ding. Responsibilities include:

  • Solve supervised causal discovery problem based on transformer neural network architecture.
  • Explore better solutions and submit two papers to a top conference as first author and co-author separately.
 
 
 
 
 
China International Capital Corporation Limited
Research Intern
China International Capital Corporation Limited
August 2023 – September 2023 Shanghai

Responsibilities include:

  • Reproduce factor mining techniques based on deep neural networks.
  • Use Bayesian neural networks to model factor uncertainty and self-attention mechanisms to model relationships between stocks.

Academic Service

  • Reviewer of NeurIPS 2024, CVPR 2024, ICML 2024, ICLR 2023, NeurIPS 2023.
  • Reviewer of IEEE Computational Intelligence Magazine.

Contests

I won the first place among 222 teams in the 2020-11 season in the graph neural network node classification competition organized by Baidu.
See certificate

Awards

  • Stars of Tomorrow (Award of Excellent Intern of Microsoft Research Asia) (2023)
  • CETC The 14TH Research Institute Glarun Scholarship (2021)
  • Three Good Student of SJTU (2018, 2020, 2021)
  • Excellent League Member of SJTU (2016, 2020, 2021)
  • Outstanding graduate of SJTU (2019)
  • Excellent Student Cadre of SJTU (2016)

Others

Contact