PhD Student

Theoretical Computer Science

Hey! I am Ziad (he/him), a current doctoral student at the University of Liverpool since October 2024. I am a member of the ACTO and NDC groups. I work under the guidance of Sebastian Wild, Nikhil Mande and Viktor Zamaraev. Prior to this, I completed my Master’s degree at the University of York under the guidance of Detlef Plump.

Note: My surname is “Ismaili Alaoui,” in full. My first name is “Ziad.” I do not have a middle name.

Interests: graph transformation, graph rewriting, graph algorithms, succinct data structures, compression, tournaments, query complexity (deterministic, randomised and quantum)…

Education

University of Liverpool

• PhD, Computer Science (Since October 2024)

University of York

• MScR, Computer Science (September 2023 to September 2024) [Thesis]
• BSc, Computer Science (September 2020 to July 2023)

Work Experience

Graduate Teaching Assistant at the University of Liverpool (January 2025 to May 2025)

• I assisted with delivering the tutorials for the module titled “Complexity of Algorithms” (COMP202).
• I assisted with demonstrating and delivering the tutorials for the module titled “Data Structures and Algorithms” (COMP108).

Graduate Teaching Assistant at the University of York (February 2024 to May 2024)

• I demonstrated for the module titled “Object Oriented Data Structures & Algorithms” (COM00016C).

Publications

[Google Scholar] [dblp]

1. Space-Efficient Hierholzer: Eulerian Cycles in O(m) Time and O(n) Space
   Ziad Ismaili Alaoui, Detlef Plump, Sebastian Wild
   Preprint (2025).
   [arXiv]

2. Succinct Preferential-Attachment Graphs
   Ziad Ismaili Alaoui, Namrata, Sebastian Wild
   WG 2025: International Workshop on Graph-Theoretic Concepts in Computer Science 2025.
   [arXiv]

3. Hardness of Finding Kings and Strong Kings
   Ziad Ismaili Alaoui, Nikhil S. Mande
   Preprint (2025).
   [arXiv]

4. Rule-Based Graph Programs Matching the Time Complexity of Imperative Algorithms
   Ziad Ismaili Alaoui, Detlef Plump
   Under review for Logical Methods in Computer Science (LMCS).
   [arXiv]

5. Linear-Time Graph Programs without Preconditions
   Ziad Ismaili Alaoui, Detlef Plump
   GCM 2024: Proceedings of the 15th International Workshop on Graph Computation Models. Electronic Proceedings in Theoretical Computer Science.
   [arXiv] [White Rose]

6. Linear-Time Graph Programs for Unbounded-Degree Graphs
   Ziad Ismaili Alaoui, Detlef Plump
   ICGT 2024: Proceedings of the 17th International Conference on Graph Transformation. Lecture Notes in Computer Science 14774, pages 3-20. Springer, 2024.
   [Springer]

Talks

1. “Succinct Preferential-Attachment Graphs”, 51st International Workshop on Graph-Theoretic Concepts in Computer Science (WG), Europäische Akademie Otzenhausen (Germany), June 12th 2025.
2. “Implementing Binary Search Trees in GP 2”, 16th International Workshop on Graph Computation Model (GCM), Koblenz (Germany), June 10th 2025.
3. “Succinct Preferential-Attachment Graphs”, Analysis of Algorithms (AofA) 2025, Toronto (Canada), May 8th 2025.
4. “Finding Strong Kings in Tournaments”, Postgraduate Combinatorics Conference (PCC) 2025, Glasgow (the United Kingdom), May 2nd 2025.
5. “Linear-Time Graph Programs for Unbounded-Degree Graphs”, International Conference on Graph Transformation (ICGT) 2024, Enschede (the Netherlands), July 11th 2024.

Random Projects

• [GP 2 (contributor)] GP 2 is a visual, rule-based graph programming language. In a nutshell, it works by algorithmically transforming an input graph into an output graph by sequentially applying graph transformation rules defined by the programmer (given in the form of a GP 2 program). I had been a contributor to the compiler while I pursued my Master’s degree at the University of York, and helped redesign and reprogram certain aspects of the compiler to enable more optimised implementations of classical graph algorithms. To give an example, the time complexity of the best-known implementation of a graph depth-first search in GP 2 was quadratic in the size of the input graph; however, the changes brought to the data structures generated by the compiler allowed for much better implementations of the algorithm, to then run in time linear to the size of the input graph! If interested, see my journal publication with Detlef Plump for a broad summary of the changes, with a few case studies (e.g. 2-colouring, acyclicity verification, or computing shortest paths).

• [The Weasal Program] A proposed C++ implementation of natural selection as per Richard Dawkins in The Blind Watchmaker. In his famous book, he explains a simple yet powerful algorithmic way to illustrate that natural process. The goal is to start with a string of random characters and mutate it until a target sentence is reached. A graphical interface for the program is available here.

• [The Mollab Project] A tentative high-school project where a user inputs the structure of a molecule and the program automatically computes its nomenclature (i.e. name). Unfortunately, the range of molecules it can support is pretty narrow.

• [Manaloo.com (discontinued)] A little recreational website consisting of a series of mathematical problems meant to be solved with a single expression/formula (probably my most exciting (ex-)project). It is relatively close to Project Euler, but instead of expecting a simple integer as an answer, the website expects a function instead (i.e. a variable expression) that it then evaluates against the expected expression. The site was sadly discontinued in 2022/23 due to high hosting fees and its small size, but it had 30+ registered members from all over the world!

---

Theme's Source