About me

I’m a 5th-year Ph.D. student at the University of California, Santa Barbara, working under the supervision of Prof. Arpit Gupta in the Systems and Networking Lab.

My current research lies at the intersection of networking and machine learning, especially network transformers and foundation models. I am interested in developing credible ML-based artifacts and data representations for different networking and security problems, making these models explainable and reliable, and democratizing network research by creating public measurement infrastructures, datasets, and tools.

I’m developing netFound - a network foundation model, designed to work with network packet traces (basically, tabular sequential hierarchical and multi-modal data) and pretrained on a massive (TBs scale) network dataset collected from real-world networks. I’m also working on demystifying and explaining current Network Foundation Models (see Demystifying Network Foundation Models) and uncovering their failure modes and problems in the benchmarking scenarios.

I’m leading netUnicorn and PINOT projects designed to simplify developing trustworthy ML solutions in networking and make real-world data available for everyone (see PINOT and netunicorn).

I also researched a Machine Learning models’ credibility in the Networking area, exploring vulnerabilities and problems in published existing solutions and how to fix them using Explainable AI methods (see TrusteeML).

Interests

• Machine Learning in Networking
  • + Network Foundation Models
  • + Explainability and robustness
  • + Model generalizability and credibility
  • + Datasets collection and curation
• Network measurement tools and infrastructures

Selected Publications

For a full list of publications, see: Google Scholar

1. In Search of netUnicorn: A Data-Collection Platform to Develop Generalizable ML Models for Network Security Problems
   Roman Beltiukov, Wenbo Guo, Arpit Gupta, Walter Willinger
   Proceedings of the 2023 ACM SIGSAC Conference on Computer and Communications Security (CCS ’23) (Nov 2023)
   Project Website
2. PINOT: Programmable Infrastructure for Networking
   Roman Beltiukov, Sanjay Chandrasekaran, Arpit Gupta, Walter Willinger
   ANRW 23: Proceedings of the Applied Networking Research Workshop (Jul 2023)
   Project Website
3. AI/ML and Network Security: The Emperor has no Clothes
   A. S. Jacobs, R. Beltiukov, W. Willinger, R. A. Ferreira, A. Gupta, L. Z. Granville
   CCS '22: Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications Security (Nov 2022)
   Project Website

Invited Talks and Tutorials

• ACM SIGCOMM 2023 Tutorial: Closed-Loop “ML for Networks” Pipelines
  ACM SIGCOMM 2023, New York, NY (09/23)
• netUnicorn: A Unified and Modular Data-Collection Platform for Developing Credible ML Models for Networking
  • UC Santa Barbara (05/22)
  • The University of Chicago (10/22)
  • ACM SIGMETRICS Workshop on Measurements for Self-Driving Networks (06/23)

Awards

• UCSB CS Department Summer Fellowship Award, 2023

Education

• Ph.D. in Computer Science – University of California, Santa Barbara 2021 - June 2026, with Prof. Arpit Gupta
• M.S. in Computer Science – Peter the Great Saint-Petersburg Polytechnical University
  (Thesis: applying natural gradient descent to reinforcement learning algorithm)
• B.S. in Computer Science – Peter the Great Saint-Petersburg Polytechnical University
  (Thesis: creating music with generative adversarial networks)

Working experience

• SWE+ Intern at Google (Summer 2025)
• Student Researcher at Google (Summer 2024)
• On-Device ML Model Researcher at Huawei Russia (Mar. 2020 - Sep. 2021)
• Junior Reinforcement Learning Researcher at JetBrains Research (Oct. 2019 - Jun. 2020)
• Python tutor at Higher School of Engineering (Sep. 2018 - Sep. 2020)

Hackathons & Challenges

• Junction 2018 – runner-up (in-team) in QOCO challenge
  (PlanPlanner – redirecting air traffic with graph algorithms)
• World-IT-Planet 2018 – winner (solo) of the international stage in Cloud Systems track
• BioHack 2018 – Special award (in-team) by Institute of Bioinformatics
  (predicting 3D cell structure with ML)
• Junction 2017 – winners (in-team) of track “Robots, Learning machines”
  (Mirror – applied ML to Pepper robot to make him copy all our movements)
• World-IT-Planet 2017 – winner (solo) of the international stage in Cloud Systems track
• Make your university 20.35 – winners (in-team)
  (recommender system for people looking for a job)

Certifications

• Microsoft Certified: Azure AI Engineer Associate, Microsoft
• Certified Associate in Python programming, OpenEDG
• Microsoft Certified Solutions Expert: Cloud Platform and Infrastructure, Microsoft
• Microsoft Certified Solutions Associate: Windows Server 2016, Microsoft
• Microsoft Certified Professional, Microsoft
• Cisco CCNA Instructor, Cisco
• Cisco Certified Network Associate Routing and Switching, Cisco
• Huawei Certified Network Associate Storage, Huawei

Thanks for your attention

Have any questions? Feel free to contact me in any way convenient for you.