Petra Wolf

PhD Student in Theoretical Computer Science

I am currently working at the University of Trier as a researcher in the Theoretical Computer Science group led by Prof. Dr. Henning Fernau.

I am interested in Quantum Computing, Formal Languages, Complexity Theory,  Computability, and Automata Theory.


Paper accepted at IJCAI 2021

Computer mit Graf

Diversity in Kemeny Rank Aggregation: A Parameterized Approach

Emmanuel Arrighi,
Henning Fernau,

Daniel Lokshtanov,

Mateus de Oliveira Oliveira,

Petra Wolf


List of Publications

[PV] = Link to Publisher's Version
[Pre] = Link to Preprint Version

[Talk] = Link to presentation

Accepted at DLT 2021 [Pre]

Properties of Graphs Specified by a Regular Language
(Volker Diekert, Henning Fernau, Petra Wolf)

Traditionally, graph algorithms get a single graph as input, and then they should decide if this graph satisfies a certain property Φ. What happens if this question is modified in a way that we get a possibly infinite family of graphs as an input, and the question if is there exists one graph satisfying Φ? We approach this question by using formal languages for specifying families of graphs. In particular, we show that certain graph properties can be decided by studying the syntactic monoid of the specification language L if a certain torsion condition is satisfied. This condition holds trivially if L is regular.
In order to show our results, we split L into a finite union of subsets Lα. Every Lα defines in a natural way a single finite graph Fα where some edges and vertices are marked. The marked graph in turn defines a set of graphs with a geometric description using the notion of graph retraction and where Fα appears as an induced subgraph.

Accepted at IJCAI 2021

Diversity in Kemeny Rank Aggregation: A Parameterized Approach
(Emmanuel Arrighi, Henning Fernau, Daniel Lokshtanov, Mateus de Oliveira Oliveira, Petra Wolf)

In its most traditional setting, the main concern of optimization theory is the search for optimal solutions for instances of a given computational problem. A recent trend of research in artificial intelligence, called solution diversity, has focused on the development of notions of optimality that may be more appropriate in settings where subjectivity is essential. The idea is instead of aiming at the development of algorithms that output a single optimal solution, the goal is to investigate algorithms outputting a small set of sufficiently good solutions that are sufficiently diverse from one another. This way, the user has the opportunity to choose the solution being most appropriate to the context at hand. It also displays the richness of the solution space.
When combined with techniques from parameterized complexity theory, the paradigm of diversity of solutions offers a powerful algorithmic framework to address problems of practical relevance. In this work, we investigate the impact of this combination in the field of Kemeny Rank Aggregation, a well-studied class of problems lying in the intersection of order theory and social choice theory and also in the field of order theory itself. In particular, we show that the Kemeny Rank Aggregation problem is fixed-parameter tractable with respect to natural parameters providing natural formalizations of the notions of diversity and of the notion of a sufficiently good solution. Our main results work both when considering the traditional setting of aggregation over linearly ordered votes, and in the more general setting where votes are partially ordered.

DCC 2021 [PV] [Talk] [Pre]

Improving Run Length Encoding by Preprocessing
(Sven Fiergolla, Petra Wolf)

The Run Length Encoding (RLE) compression method is a long standing simple lossless compression scheme which is easy to implement and achieves a good compression on input data which contains repeating consecutive symbols. In its pure form RLE is not applicable on natural text or other input data with short sequences of identical symbols. We present a combination of preprocessing steps that turn arbitrary input data in a byte-wise encoding into a bit-string which is highly suitable for RLE compression. The main idea is to first read all most significant bits of the input byte-string, followed by the second most significant bit, and so on. We combine this approach by a dynamic byte remapping as well as a Burrows-Wheeler-Scott transform on a byte level. Finally, we apply a Huffman Encoding on the output of the bit-wise RLE encoding to allow for more dynamic lengths of code words encoding runs of the RLE. With our technique we can achieve a lossless average compression which is better than the standard RLE compression by a factor of 8 on average.

FSTTCS 2020 [PV] [Talk]

Width Notions for Ordering-Related Problems
(Emmanuel Arrighi, Henning Fernau, Mateus de Oliveira Oliveira, Petra Wolf)

We are studying a weighted version of a linear extension problem, given some finite partial order ρ, called Completion of an Ordering. While this problem is NP-complete, we show that it lies in FPT when parameterized by the interval width of ρ. This ordering problem can be used to model several ordering problems stemming from diverse application areas, such as graph drawing, computational social choice, or computer memory management. Each application yields a special ρ. We also relate the interval width of ρ to parameterizations such as maximum range that have been introduced earlier in these applications, sometimes improving on parameterized algorithms that have been developed for these parameterizations before. This approach also gives some practical sub-exponential time algorithms for ordering problems.

FSTTCS 2020 [PV] [Talk] [Pre]

Synchronization under Dynamic Constraints
(Petra Wolf)

Imagine an assembly line where a box with a lid and liquid in it enters in some unknown orientation. The box should leave the line with the open lid facing upwards with the liquid still in it. To save costs there are no complex sensors or image recognition software available on the assembly line, so a reset sequence needs to be computed. But how can the dependencies of the deforming impact of a transformation of the box, such as 'do not tilt the box over when the lid is open' or 'open the lid again each time it gets closed' be modeled? We present three attempts to model constraints of these kinds on the order in which the states of an automaton are transitioned by a synchronizing word. The first two concepts relate the last visits of states and form constraints on which states still need to be reached, whereas the third concept concerns the first visits of states and forms constraints on which states might still be reached. We examine the computational complexity of different variants of the problem, whether an automaton can be synchronized with a word that respects the constraints defined in the respective concept, and obtain nearly a full classification. While most of the problems are PSPACE-complete we also observe NP-complete variants and variants solvable in polynomial time.
One of them is the careful synchronization problem for partial weakly acyclic automata (which are partial automata, the states of which can be ordered such that no transition leads to a smaller state),    which is shown to be solvable in time O(k^2 n^2 log(n)) where n is the size of the state set and k is the size of the alphabet. The algorithm even computes a synchronizing words as a witness. This is quite surprising as the careful synchronization problem uses to be a hard problem for most classes of automata.
We will also observe a drop of the complexity if we track the orders of states on several paths simultaneously instead of tracking the set of active states. Further, we give upper bounds on the length of a synchronizing word depending on the size of the input relation and show that the Cerny conjecture holds for partial weakly acyclic automata.

FSTTCS 2020 [PV] [Talk] [Pre]

Synchronization of Deterministic Visibly Push-Down Automata
(Henning Fernau, Petra Wolf)

We generalize the concept of synchronizing words for finite automata, which map all states of the automata to the same state, to deterministic visibly push-down automata. Here, a synchronizing word w does not only map all states to the same state but also fulfills some conditions on the stack content of each run after reading w. We consider three types of these stack constraints: after reading w, the stack (1) is empty in each run, (2) contains the same sequence of stack symbols in each run, or (3) contains an arbitrary sequence which is independent of the other runs. We show that in contrast to general deterministic push-down automata, it is decidable for deterministic visibly push-down automata whether there exists a synchronizing word with each of these stack constraints, i.e., the problems are in EXPTIME. Under the constraint (1) the problem is even in P. For the sub-classes of deterministic very visibly push-down automata the problem is in P for all three types of constraints. We further study variants of the synchronization problem where the number of turns in the stack height behavior caused by a synchronizing word is restricted, as well as the problem of synchronizing a variant of a sequential transducer, which shows some visibly behavior, by a word that synchronizes the states and produces the same output on all runs.

MFCS 2020 [PV] [Talk] [Pre]

Synchronizing Deterministic Push-Down Automata Can Be Really Hard
(Henning Fernau, Petra Wolf, Tomoyuki Yamakami)

The question if a deterministic finite automaton admits a software reset in the form of a so-called synchronizing word can be answered in polynomial time. In this paper, we extend this algorithmic question to deterministic automata beyond finite automata. We prove that the question of synchronizability becomes undecidable even when looking at deterministic one-counter automata. This is also true for another classical mild extension of regularity, namely that of deterministic one-turn push-down automata. However, when we combine both restrictions, we arrive at scenarios with a PSPACE-complete (and hence decidable) synchronizability problem. Likewise, we arrive at a decidable synchronizability problem for (partially) blind deterministic counter automata.
There are several interpretations of what synchronizability should mean for deterministic push-down automata. This is depending on the role of the stack: should it be empty on synchronization, should it be always the same or is it arbitrary? For the automata classes studied in this paper, the complexity or decidability status of the synchronizability problem is mostly independent of this technicality, but we also discuss one class of automata where this makes a difference.

ICTCS 2020 [PV] [Pre]

From Decidability to Undecidability by Considering Regular Sets of Instances
(Petra Wolf)

We are lifting classical problems from single instances to regular sets of instances. The task of finding a positive instance of the combinatorial problem P in a potentially infinite given regular set is equivalent to the so called intreg-problem of P, which asks for a given DFA A, whether the intersection of P with L(A) is non-empty. The intreg-problem generalizes the idea of considering multiple instances at once and connects classical combinatorial problems with the field of automata theory. While the question of the decidability of the intreg-problem has been answered positively for several NP- and even PSPACE-complete problems, we are presenting natural problems even from L with an undecidable intreg-problem. We also discuss alphabet sizes and different encoding-schemes elaborating the boundary between problem-variants with a decidable respectively undecidable intreg-problem.

arXiv [Pre]

Regular Intersection Emptiness of Graph Problems: Finding a Needle in a Haystack of Graphs with the Help of Automata
(Heinning Fernau, Petra Wolf)

The Int_reg-problem of a combinatorial problem P asks, given a nondeterministic automaton M as input, whether the language L(M) accepted by M contains any positive instance of the problem P. We consider the Int_reg-problem for a number of different graph problems and give general criteria that give decision procedures for these Int_reg-problems. To achieve this goal, we consider a natural graph encoding so that the language of all graph encodings is regular. Then, we draw the connection between classical pumping- and interchange-arguments from the field of formal language theory with the graph operations induced on the encoded graph. Our techniques apply among others to the Int_reg-problem of well-known graph problems like Vertex Cover and Independent Set, as well as to subgraph problems, graph-edit problems and graph-partitioning problems, including coloring problems.

MFCS 2019 [PV]

Computational Complexity of Synchronization under Regular Constraints
(Henning Fernau, Vladimir V. Gusev, Stefan Hoffmann, Markus Holzer, Mikhail V. Volkov, Petra Wolf)

Many variations of synchronization of finite automata have been studied in the previous decades. Here, we suggest studying the question if synchronizing words exist that belong to some fixed constraint language, given by some partial finite automaton called constraint automaton. We show that this synchronization problem becomes PSPACE-complete even for some constraint automata with two states and a ternary alphabet. In addition, we characterize constraint automata with arbitrarily many states for which the constrained synchronization problem is polynomial-time solvable. We classify the complexity of the constrained synchronization problem for constraint automata with two states and two or three letters completely and lift those results to larger classes of finite automata.

DCFS 2019 [PV]

On the Decidability of Finding a Positive ILP-Instance in a Regular Set of ILP-Instances
(Petra Wolf)

The regular intersection emptiness problem for a decision problem P (intreg(P)) is to decide whether a potentially infinite regular set of encoded P-instances contains a positive one. Since intreg(P) is decidable for some NP-complete problems and undecidable for others, its investigation provides insights in the nature of NP-complete problems. Moreover, the decidability of the intreg -problem is usually achieved by exploiting the regularity of the set of instances; thus, it also establishes a connection to formal language and automata theory. We consider the intreg-problem for the well-known NP-complete problem Integer Linear Programming (ILP). It is shown that any DFA that describes a set of ILP-instances (in a natural encoding) can be reduced to a finite core of instances that contains a positive one if and only if the original set of instances did. This result yields the decidability of  intreg(ILP).

FUN 2018 [PV]

We study the game Greedy Spiders, a two-player strategic defense game, on planar graphs and show PSPACE-completeness for the problem of deciding whether one player has a winning strategy for a given instance of the game. We also generalize our results in metatheorems, which consider a large set of strategic defense games. We achieve more detailed complexity results by restricting the possible strategies of one of the players, which leads us to Sigma^p_2- and Pi^p_2-hardness results.

LATA 2018 [PV]

For a regular set R of quantified Boolean formulae we decide whether R contains a true formula. We conclude that there is a PSPACE-complete problem for which emptiness of intersection with a regular set is decidable. Furthermore, by restricting depth and order of quantification we obtain complete problems for each level of the polynomial hierarchy with this decidability as well.

Bachelor and Master Thesis

Bachelor Thesis (12.2016)

Computational Complexity of Computer Games - Gaming under Time Pressure

Identification of combination of game elements which give a game, in which they
occur, a certain computational complexity. Abstracted game elements have been studied instead of
concrete games. The work focused on the stopwatch as a game element.

[Download (German)]

Master Thesis (10.2018)

Decidability of the Regular Intersection
Emptiness Problem

We determine the decidability of the intreg-problem for several languages. The intreg-problem of a fixed language L asks whether the intersection of L with a given regular language R is empty or not. We prove undecidability of the intreg-problem for variations of the Machine Language, Bounded Tiling, Corridor Tiling, Bounded PCP, Equivalence of Regular Expressions in a shuffled encoding, and String Equivalence Modulo Padding in a shuffled encoding. We prove decidability of the intreg-problem for String Equivalence Modulo Padding in a sequential encoding as well as over a unary alphabet in both encodings, SAT, k-TQBF, True Quantified Boolean Formula, Integer Linear Programming, Vertex Cover, Independent Set, Knapsack, and Integer Knapsack. In the most cases we show the decidability of the intreg-problem by constructing a condensed automaton condensed(A) from a given DFA A. In L(condensed(A)) only finitely many words have to be tested for membership in the intersection (defined by intreg) under which there is one element in the intersection if and only there is one in the original regular language described by A. We develop the three techniques merging, separating, and replacing of constructing the automaton  condensed(A) and demonstrate them on the problems Vertex Cover, Independent Set, and Knapsack.

[Download (English)]



Eberhard Karls Universität Tübingen

  • Bachelor of Science in Computer Science 2016

  • Master of Science in Computer Science 2018

Universität Trier

In progress: PhD supervised by Prof. Dr. Henning Fernau

Scholarships and Awards

  • Master degree with distinction (Master Abschluss mit Auszeichnung) 2018

  • Deutschlandstipendium 2017 - 2019

  • JSPS Summer Program 2020

  • Publication award of the graduate college of University Trier Faculty Four 2021
    (Publikationspreis im Fachbereich 4 des Graduiertenkollegs der Universität Trier) 2021

Lehrer eine Formel auf einer Tafel schre


Exercises/Course Administration/Lecturer

Universität Trier:

  • 2021

    • Ausgewählte Themen zu Quantum-Computing

  • 2020/2021

    • Vorkurs Formale Grundlagen der Informatik

    • Diskrete Strukturen

    • Reading Group: Quantum Computing

  • 2019, 2020, 2021 Kleines Studienprojekt LaTeX-Kurs

Eberhard Karls Universität Tübingen:

  • 2018

    • Proseminar Komplexität von Computerspielen

  • 2017/2018

    • Theoretische Informatik

    • Proseminar Theoretische Informatik - Große Informatiker der Geschichte

  • 2017

    • Berechenbarkeit

    • Formale Sprachen

    • Proseminar Theoretische Informatik - Reguläre Sprachen

  • 2016/2017

    • Theoretische Informatik

    • Mathematischer Vorkurs

  • 2016

    • Komplexitätstheorie

  • 2015/2016

    • Theoretische Informatik


Student Supervision

RP = Research Project

BA = Bachelor Thesis,

MA = Master Thesis,

  • Jonas Lingg (RP) 2020: The Consistency Problem for Partially Ordered Automata and Permutation Automata
  • Sven Fiergolla (BA) 2020: Improving Run Length Encoding on bit level by preprocessing
  • Alexander Pet (BA) 2021: Implementierung des historischen Brettspiels Latrunculi mit einer künstlichen Intelligenz in Unity
  • Kevin Goergen (BA) 2021: Computational Complexity of the Puzzle Game Roma
Keyboard and Mouse
Black Notebook


Petra Wolf

Theoretische Informatik

Campus II

Universität Trier
D-54286 Trier



„Computer Science is no more about computers than astronomy is about telescopes.“

Edsger Wybe Dijkstra