Meisam Mohammady

Meisam Mohammady

Address: 2434 Osborn Dr, Department of Computer Science, Iowa State University, Ames, IA, 50011
Office: 232 Atanasoff Hall
E-mail: meisam@iastate.edu

Research

Differential Privacy, Secure Federated Learning, Anonymity, Computational Learning Theory, Secure Multiparty Computation, Fairness

Awarded Grants

Data 61 PhD Scholarship Grant

“Sub-optimal but Comprehensive Approach for AI with Differential Privacy and Fairness”

  • Role: PI.
  • Project Duration: 08/01/2021-08/31/2025.
  • Awarded Amount: $55,000 per annum.

Vacation Students Scholarship Grant “Utility-driven Statistical Inference Engine with Local Differential Privacy”

  • Role: PI.
  • Project Duration: 12/01/2021-03/31/2022.
  • Awarded Amount: $22,000 (CSIRO Data 61, Pawsey Supercomputing)

Invited Talk

  1. Preserving Both Privacy and Utility in Network Trace Anonymization”, Universit ́e du Qu ́ebec `a Montr ́eal (UQAM), Montr ́eal, Canada, November 22, 2019
  2. R2DP: A Universal Approach to Optimizing the Randomization Mechanisms of Differential Privacy for Utility Metrics with No Known Optimal Distributions”, Universit ́e du Qu ́ebec `a Montr ́eal (UQAM), Montr ́eal, Canada, November 22, 2019
  3. DP-IDS: Differentially Private Intrusion Detection System ”, Security, Privacy and Forensics (SPF) seminars, Montr ́eal, Canada, May 10, 2019
  4. R2DP: A Universal Approach to Optimizing the Randomization Mechanisms of Differential Privacy for Utility Metrics with No Known Optimal Distributions”, The CSIRO, Data61 Reading seminar, Sydney, Australia, November 22, 2020
  5. Novel Approaches to Preserving Utilityin Privacy Enhancing Technologies,Discovery Partners Institute (DPI) RD Seminar, Chicago, IL, USA, September 9, 2021

Demonstrations

  1. Preserving Both Privacy and Utility in Network Trace Anonymization”, Ericsson Security Research, Montr ́eal, Canada, May, 2018
  2. R2DP: A Universal and Automated Approach to Optimizing the Randomization Mechanisms of Differential Privacy for Utility Metrics with No Known Optimal Distributions”, Ericsson Security Research, Montr ́eal, Canada, October, 2019
  3. DPOAD: Differentially Private Outsourcing of Anomaly Detection with Optimal Sensitivity Learning”, Ericsson Security Research, Montr ́eal, Canada, October, 2020