Arman Riasi

Ph.D. Candidate at Virginia Tech · Visiting Scholar at Purdue University

I am a Ph.D. candidate in Computer Science at Virginia Tech and a visiting scholar at Purdue University. My research focuses on trustworthy AI, particularly verifiable computation (e.g., zero-knowledge proofs), privacy-preserving AI/ML, and federated learning.

Arman Riasi

Education

M.S. in Computer Science, Virginia Tech 2025
B.S. in Computer Engineering, Isfahan University of Technology 2022

Experience

Visiting Scholar, TCLS Lab, Purdue University Aug 2026–Present

Host: Prof. Zahra Ghodsi

West Lafayette, IN

Graduate Research Assistant, VT-ASAP Lab, Virginia Tech Aug 2022–Present

Blacksburg, VA

AI Agents and Data Support Intern, Virginia Tech Jul 2026–Aug 2026

Blacksburg, VA

Graduate Teaching Assistant, Virginia Tech Aug 2022–May 2026

Blacksburg, VA

Publications

2025

Trustworthy Federated Learning with Local Differential Privacy

Rouzbeh Behnia, Jeremiah Birrell, Arman Riasi, Reza Ebrahimi, Kaushik Dutta, and Thang Hoang

Workshop on Information Technologies and Systems (WITS)

Best Paper Award

Paper DOI

Zero-Knowledge AI Inference with High Precision

Arman Riasi, Haodi Wang, Rouzbeh Behnia, Viet Vo, and Thang Hoang

ACM Conference on Computer and Communications Security (ACM CCS)

Acceptance rate: 13.9%

Paper Code DOI

2024

Privacy-Preserving Verifiable Neural Network Inference Service

Arman Riasi, Jorge Guajardo, and Thang Hoang

Annual Computer Security Applications Conference (ACSAC)

Acceptance rate: 19.7%

Paper Code DOI

Efficient Secure Aggregation for Privacy-Preserving Federated Machine Learning

Rouzbeh Behnia, Arman Riasi, Mohammadreza Ebrahimi, Sherman S. M. Chow, Balaji Padmanabhan, and Thang Hoang

Annual Computer Security Applications Conference (ACSAC)

Acceptance rate: 19.7%

Paper Code DOI

Breaking Privacy in Model-Heterogeneous Federated Learning

Atharva Haldankar, Arman Riasi, Hoang-Dung Nguyen, Tran Phuong, and Thang Hoang

International Symposium on Research in Attacks, Intrusions and Defenses (RAID)

Acceptance rate: 25%

Paper Code DOI

Patents

Method and System for Next-Location Prediction Using Federated Learning with Heterogeneous Mobility Data

Arman Riasi, Rouzbeh Behnia, Thang Hoang, and Kaushik Dutta

U.S. Provisional Patent Application No. 63/969,801

Filed January 28, 2026 · Inventor

Awards

Best Paper Award, WITS 2025 (selected from 156 papers) Dec 2025
CCI SWVA Cyber Innovation Scholarship Mar 2025, 2026
ACSAC Student Conferenceship Grant (NSF-Funded) Oct 2024
BitShares Graduate Student Fellowship June 2023

Professional Service

Artifact Evaluation Program Committee

  • PETS, 2027
  • NDSS, 2026–2027
  • ACM CCS, 2025–2026
  • USENIX Security, 2026

Poster & Demo Program Committee

  • ACM CCS, 2026

Journal Reviewer

  • IEEE TNSE, 2026
  • IEEE TDSC, 2024–2026
  • IEEE TCC, 2023–2025

Conference Reviewer

  • ACM CCS, 2026
  • IEEE S&P, 2026
  • ACISP, 2026
  • ACM AsiaCCS, 2025
  • PETS, 2023–2025
  • ACSAC, 2023–2025