Arion Scheid

M.Sc. Logic and AI · TU Wien

Arion Scheid

Probabilistic program verification, rewriting, and the logic behind machine learning.

Vienna, Austria

I am a Master’s student in Logic and AI at TU Wien in Vienna. My interests lie in probabilistic program verification, rewriting, explainable AI, theoretical machine learning and logic, and more broadly in understanding when and why automated systems can be trusted.

Before moving to Vienna, I completed my B.Sc. in Informatics at RWTH Aachen University, including an exchange semester in the Formal Methods master’s programme at Universidad Complutense de Madrid. In my bachelor’s thesis at the Programming Languages and Verification group, I showed how stochastic context-free grammars can be used to decide almost-sure termination for a class of probabilistic term rewriting systems.

  • Probabilistic Program Verification
  • Rewriting
  • Explainable AI
  • Theoretical Machine Learning
  • Logic
01

CV

Education

  1. 2026 – present

    M.Sc. Logic and AI

    TU Wien · Vienna, AT

  2. 2020 – 2025
  3. Winter 2024

    Exchange Semester, M.Sc. Formal Methods in Computer Science

    Universidad Complutense de Madrid · Madrid, ES

Teaching

  1. Apr 2022 – Aug 2024

    Student Tutor

    RWTH Aachen University · Aachen, DE

    • Foundations of Logic Programming
    • Analysis for Computer Scientists
    • Introduction to Programming
    • Data Structures and Algorithms
    • Formal Languages, Automata and Processes
Full CV (PDF)
02

Publications

2026

2025

  1. Using Stochastic Context-Free Grammars to Decide Almost-Sure Termination for Probabilistic Term Rewriting Systems

    Arion Scheid

    ThesisBachelor’s Thesis, RWTH Aachen University

03

Projects

  • Vienna District Browser

    2026

    A knowledge graph project built for a course at TU Wien, modelling and exploring information about Vienna’s districts.

    • Knowledge Graphs
    • Python
  • GNNExplainer: Seminar Study

    2026

    Seminar work for “Theoretical Aspects of Machine Learning” at TU Wien, examining Ying et al.’s 2019 paper GNNExplainer: Generating Explanations for Graph Neural Networks, and a fatal flaw in the theory the paper rests upon.

    • Explainable AI
    • Graph Neural Networks
    • LaTeX
  • Visual Learning Tool

    2026

    A visual learning tool. Coming soon…

04

Contact

The best way to reach me is by email. I'm always happy to talk about research, collaborations or anything else. You can also find me on the platforms below.