Selected Publication [full list]

Preprints

• Why Is Public Pretraining Necessary for Private Model Training?
  Arun Ganesh, Mahdi Haghifam, Milad Nasr, Sewoong Oh, Thomas Steinke, Om Thakkar, Abhradeep Thakurta, Lun Wang. (Authors are ordered alphabetically.)
  
• FED-χ2: Privacy Preserving Federated Correlation Test
  Lun Wang*, Qi Pang*, Shuai Wang, Dawn Song.
  Preliminary version appeared at PPAI Workshop @ AAAI 2022.

2023

• Byzantine-Robust Federated Learning with Optimal Rates and Privacy Guarantee [code]
  Banghua Zhu*, Lun Wang*, Qi Pang*, Shuai Wang, Jiantao Jiao, Dawn Song, Michael I.Jordan.
  AISTATS 2023: Artificial Intelligence and Statistics 2023. (Acceptance rate: 29%)
  SpicyFL @ Neurips 2020: NeurIPS Workshop on Scalability, Privacy, and Security in Federated Learning.

2022

• Differentially Private Fractional Frequency Moments Estimation with Polylogarithmic Space
  Lun Wang, Iosif Pinelis, Dawn Song.
  ICLR 2022: the 10th International Conference on Learning Representations. (Acceptance rate: 1095/3391=32.3%))
  PPAI Workshop @ AAAI 2022: The 3rd AAAI Workshop on Privacy-Preserving Artificial Intelligence.
  TPDP Workshop @ ICML 2022: Theory and Practice of Differential Privacy.
• PRIVGUARD: Privacy Regulation Compliance Made Easier
  Lun Wang, Usmann Khan, Joseph Near, Qi Pang, Jithendaraa Subramanian, Neel Somani, Peng Gao, Andrew Low, Dawn Song.
  Usenix Security'22: the 31th Usenix Security Symposium.

2021

• BACKDOORL: Backdoor Attack against Competitive Reinforcement Learning
  Lun Wang, Zaynah Javed, Xian Wu, Wenbo Guo, Xinyu Xing, Dawn Song.
  IJCAI'21: the 30th International Joint Conference on Artificial Intelligence. (Acceptance rate: 587/4204=13.9%)

2020

• Towards practical differentially private causal graph discovery. [code]
  Lun Wang, Qi Pang, Dawn Song.
  NeurIPS'20: Proceedings of The 34th Annual Conference on Neural Information Processing Systems. (Acceptance rate: 1900/9454=20.1%)
• Towards Inspecting and Eliminating Trojan Backdoors in Deep Neural Networks.
  Wenbo Guo*, Lun Wang*, Yan Xu, Xinyu Xing, Min Du, Dawn Song.
  ICDM'20: 20th IEEE International Conference on Data Mining. (Acceptance rate: 91/930=9.8%)

2019

• CHURP: Dynamic-committee proactive secret sharing. [project page]/[code]
  SKD Maram, Fan Zhang, Lun Wang, Andrew Low, Yupeng Zhang, Ari Juels, Dawn Song.
  CCS'19: Computer and Communications Security 2019. (Acceptance Rate: 149/934=16.0%)
• DUET: An expressive higher-order language and linear type system for statically enforcing differential privacy. [code]
  Joseph Near, David Darais, Chike Abuah, Tim Stevens, Pranav Gaddamadugu, Lun Wang, Neel Somani, Mu Zhang, Nikhil Sharma, Alex Shan, Dawn Song.
  OOPSLA'19: Proceedings of the ACM on Programming Languages, 2019. Distinguished Paper Award. (2.5% of all submissions)
• Towards practical differentially private convex optimization. [code]
  Roger Iyengar, Joseph Near, Dawn Song, Om Thakkar, Abhradeep Thakurta, Lun Wang. (Authors are ordered alphabetically.)
  Oakland'19: 2019 IEEE Symposium on Security and Privacy. (Acceptance rate: 12%)
• A generic technique for sketches to adapt to different counting ranges.
  Tong Yang, Jiaqi Xu, Xilai Liu, Peng Liu, Lun Wang, Jun Bi, Xiaoming Li.
  INFOCOM'19: IEEE Conference on Computer Communications.

• Reviewer: AISTATS 2022-2023, ICLR 2022-2023, Neurips 2021-2022, CCS 2021, ICML 2021-2023, WWW 2021, ECML-PKDD 2021, TDSC 2021, PoPETS 2021-2024, TLMR, ITCS 2023
• Program Committee: CCS 2023, SPAI 2020