Weekly research seminar
Organizers
- prof. dr hab. Mikołaj Bojańczyk
- prof. dr hab. Damian Niwiński
Information
Wednesdays, 2:15 p.m. , room: 5440Research fields
List of talks
-
Jan. 15, 2025, 2:15 p.m.
Mikołaj Bojańczyk (University of Warsaw)
Graphs of unbounded linear cliquewidth. (Graphs of unbounded linear cliquewidth.)
We show that for every class C of graphs, one of the following holds: 1. The class has bounded linear cliquewidth; or 2. There is a surjective MSO transduction from C to the class of …
-
Dec. 18, 2024, 2:15 p.m.
Lorenzo Clemente (University of Warsaw)
The commutativity problem for varieties of formal series and applications (The commutativity problem for varieties of formal series and applications)
Let Σ be a finite alphabet of noncommuting indeterminates. A *formal series* over the rational numbers is a mapping Σ* → ℚ. We say that it is *commutative* if the output does not depend on …
-
Dec. 11, 2024, 2:15 p.m.
Martin Grohe (RWTH Aachen University)
Some Thoughts on Graph Similarity (Some Thoughts on Graph Similarity)
We give an overview of different approaches to measuring the similarity of, or the distance between, two graphs, highlighting connections between these approaches. We also discuss the complexity of computing the distances.
-
Dec. 4, 2024, 2:15 p.m.
Paweł Parys (University of Warsaw)
A Dichotomy Theorem for Ordinal Ranks in MSO (A Dichotomy Theorem for Ordinal Ranks in MSO)
We focus on formulae ∃X.ϕ(Y, X) of monadic second-order logic over the full binary tree, such that the witness X is a well-founded set. The ordinal rank rank(X) < ω_1 of such a set X …
-
Nov. 27, 2024, 2:15 p.m.
Michał Skrzypczak (University of Warsaw)
Index problem for tree automata - revisited (Index problem for tree automata - revisited)
This talk is mostly "other people's work" type of a talk. However, it is meant to prepare ground for the next week, when Paweł Parys will present our joint results with Damian Niwiński. I want …
-
Nov. 20, 2024, 2:15 p.m.
Mikołaj Bojańczyk (University of Warsaw)
Transducers and straight line programs (Transducers and straight line programs)
In this talk, a straight line program (SLP) is a context-free grammar that generates a single string. This can be viewed as a compression scheme, with the compressed string being possibly exponentially smaller than the …
-
Nov. 13, 2024, 2:15 p.m.
Szymon Toruńczyk (University of Warsaw)
Merge-width and First-Order Model Checking (Merge-width and First-Order Model Checking)
We introduce merge-width, a family of graph parameters that unifies several structural graph measures, including treewidth, degeneracy, twin-width, clique-width, and generalized coloring numbers. Graph classes of bounded merge-width – for which the radius-r merge-width parameter …
-
Nov. 6, 2024, 2:15 p.m.
Mathieu Lehaut (University of Gothenburg)
Synthesis for first order specifications over data words (Synthesis for first order specifications over data words)
We study the reactive synthesis problem for distributed systems with a finite but unbounded number of participants interacting with an uncontrollable environment. Executions of those systems are modelled by data words, and specifications are given …
-
Oct. 30, 2024, 2:15 p.m.
Piotr Hofman (University of Warsaw)
Soundness for Workflows with Reset Arcs (Soundness for Workflows with Reset Arcs)
Workflows are a subclass of Petri nets designed to model business processes. Various definitions of soundness for workflows have been proposed, with the two most prominent being "soundness" and the stronger "generalized soundness." We begin …
-
Oct. 23, 2024, 2:15 p.m.
Antoni Puch (University of Warsaw)
Minimising ambiguity of weighted automata (Minimising ambiguity of weighted automata)
By bounding the number of accepting runs of automata we obtain various subclasses of them. We consider: unambiguous automata, finitely-ambiguous automata, and polynomially-ambiguous automata, where the number of runs is bounded by 1, a constant, …
-
Oct. 16, 2024, 2:15 p.m.
Antonio Casares (University of Warsaw)
An algorithm to rule them all: Fast value iteration for energy games (An algorithm to rule them all: Fast value iteration for energy games)
We introduce a uniform framework for describing, comparing, and proving correctness of algorithms for energy games (which generalise the well known parity games). This framework allows us to provide simple presentations and correctness proofs of …
-
Oct. 9, 2024, 2:15 p.m.
Lorenzo Clemente (University of Warsaw)
Control theory and series in noncommuting variables (Control theory and series in noncommuting variables)
We present Michel Fliess' work from the 1970's showing surprising connections between control theory and the theory of series in noncommuting variables. More precisely, to a dynamical system with inputs S one can associate a …
-
Oct. 2, 2024, 2:15 p.m.
Ioannis Eleftheriadis (University of Cambridge)
Preservation theorems on tame classes of finite structures (Preservation theorems on tame classes of finite structures)
Preservation theorems provide a correspondence between the syntactic structure and the semantic behaviour of first-order sentences. The study of preservation theorems relativised to classes of finite structures was initiated by Atserias, Dawar, and Kolaitis [JACM, …
-
Sept. 4, 2024, 2:15 p.m.
Sven Manthe (University of Bonn)
The Borel monadic theory of order is decidable (The Borel monadic theory of order is decidable)
When proving decidability of S2S, Rabin derived decidability of the monadic theory of (ℝ,<) with quantification restricted to Fσ-sets. Undecidability of the unrestricted monadic theory of (ℝ,<) was proven by Shelah. We discuss decidability for …
-
June 19, 2024, 2:15 p.m.
Teodor Knapik (ISEA, Université de la Nouvelle Calédonie)
Lindenmayer graph languages, first-order theories and expanders (Lindenmayer graph languages, first-order theories and expanders)
Imagined by Kolmogorov in the middle of past century, expanders form remarkable graph families with applications in areas as diverse as robust communication networks and probabilistically checkable proofs, to name just two. Since the proof …