Logic Seminar

The Logic seminar is held on Wednesdays, usually at 12-14. If not separately specified, the talks always start at a quarter past.

We occasionally have online talks - the permanent Zoom room for the seminar is: https://helsinki.zoom.us/j/62891400777?pwd=UldCeThTaTJVQjUzUFo4S2ErcndNQT09 (Meeting ID: 628 9140 0777, Passcode: 164195)

The seminar is led by prof. Juha Kontinen and Åsa Hirvonen.

Schedule of the spring term 2026

Wed 14.1.2026 12-14, C124

No seminar

Wed 21.1.2026 12-14, C124

Kai Sauerwald: New Results on the Complexity of Preferential Reasoning

Abstract: In this talk, we will consider recent results on the complexity of preferential reasoning in the style of Kraus, Lehmann, and Magidor. We begin with a brief overview of the approach and classical results. Novel complexity results on propositional preferential reasoning are presented. In the last part, we will then consider the combination of preferential reasoning with propositional dependence logic and the (non)feasibility of inference with this context.

Wed 28.1.2026 12-14, C124

Veeti Ahvonen (Tampere University): Logical Foundations for Modern Computational Models

Wed 4.2.2026 12-14, C124

seminar cancelled

Wed 11.2.2026 12-14, C124

Octavian Neculau: A non-structure theorem for the theory of differentially closed fields

Abstract: One of the most common ideas in all areas of mathematics is to have a way of telling whether two, or more considered objects are independent in a suitable sense. It turns out that that non-forking provides us a way of talking about independence in model theory. 

In this talk I will try to give an overview of my master's thesis, focusing on the connection between algebra and model theory, in particular on how non-forking relates to independence notions we usually define for a particular class of algebraic structures. The main result of my thesis, which will be briefly presented, states that for all regular cardinals $\kappa > \omega_1$ such that $\kappa^\omega = \kappa$, the theory of differentially closed fields has $2^\kappa$-many non-isomorphic models of cardinality $\kappa$, unlike the theory of algebraically closed fields.

Wed 18.2.2026 12-14, C124

Matilda Häggblom: Intersections in team semantics and an axiomatization of propositional inclusion-dependence logic

Abstract: We examine intersection closure in team semantics and define a logic expressively complete for all intersection-closed team properties. A novel variant of the disjunction is introduced for this purpose, and we compare it to the split-disjunction. This comparison shows that the full team in the intersection‑closed setting serves a role analogous to that of the empty team in the union‑closed setting. 

We also axiomatize propositional inclusion logic extended with dependence atoms, based on joint work with Fan Yang. The axiomatization takes on a surprisingly modular form and provides us with a method that, under certain conditions, can be used for other logics augmented with dependence atoms.

Wed 25.2.2026 12-14, C124

No seminar

Wed 4.3.2026 12-14, C124

Exam week, no seminar

Wed 11.3.2026 12-14, C124

Tuomo Lehtonen: Results and open problems for the complexity of argumentative reasoning

Abstract: Abstract rule-based argumentation (ASPIC+) is the most general structured argumentation formalism. It formalises 1) constructing arguments for and against claims from a knowledge base of premises and derivation rules, and 2) deciding which arguments are jointly acceptable, based on the inconsistencies (or "attacks") between arguments. In this talk, I focus on the complexity of reasoning in ASPIC+ when preferences between premises and rules are allowed, and are used to resolve symmetric inconsistencies.

The challenge of analyzing the complexity of reasoning in ASPIC+ lies in the fact that the number of arguments is not polynomially bounded. With coauthors, we have re-characterized ASPIC+ semantics to bypass the explicit construction of arguments, allowing for establishing complexity results for significant fragments of ASPIC+. It turns out that other fragments are quite a bit trickier. I will review the positive results and discuss the interesting challenges regarding the open problems.

The talk is based on

Lehtonen, Odekerken, Wallner, and Järvisalo: Complexity Results and Algorithms for Preferential Argumentative Reasoning in ASPIC+ (KR 2024)

Lehtonen, Wallner, and Järvisalo: Computing Stable Conclusions under the Weakest-Link Principle in the ASPIC+ Argumentation Formalism (KR 2022)

Wed 18.3.2026 12-14, C124

Manon Blanc: Robustness and (un)decidability: from Noise to Effective Topology

Reasoning about dynamical systems evolving over the reals is well-known to lead to undecidability. However, various results in the literature have shown that decision procedures exist when restricting to robust systems, provided a suitably chosen notion of robustness is employed. In particular, in verification, it has been established that if the state reachability is not sensitive to infinitesimal perturbations, then decision procedures for state reachability exist. More fundamentally, while all these statements are only about computability issues, we also consider complexity theory aspects. We prove that robustness to some precision is inherently related to the complexity of the decision procedure. Under additional hypotheses, we also prove that the reachability relation for robust systems is PSPACE-complete. We also generalise and formalise this notion of perturbation, or noise, to broader contexts in topology.

Wed 25.3.2026 12-14, C124

Ur Ya'ar: The modal logic of forcing

Abstract: The technique of forcing in set-theory allows us to take a certain model of ZFC and construct from it a new, larger model. This larger model, called a forcing extension, may satisfy different statements than the original one. We may then say that the statements holding in some forcing extension are “possible” from the point of view of the original model, while statements holding in all forcing extensions are “necessary”. This interpretation naturally suggests using modal logic to investigate the properties of the forcing technique. It also fits well with the “possible worlds” semantics for modal logic – with models of ZFC as worlds, and the accessibility relation being “is a forcing extension of...” .

In their paper “The modal logic of forcing”, Hamkins and Löwe laid the foundation for this investigation, and proved that if ZFC is consistent, then the ZFC-provable principles of forcing, i.e. modal statements ψ(q₀,...,qₙ) such that for every set-theoretic statements ϕ₀,...,ϕₙ, ZFC ⊢ ψ(ϕ₀,...,ϕₙ), are exactly the modal theory known as S4.2. Following this result many related questions arose, such as restricting our attention to a specific class of forcing notions, or focusing on the principles of forcing over a particular model of ZFC.

In this talk I will attempt to survey the state of current knowledge in the field, give a glimpse into the techniques used, and present open questions for further exploration.

Wed 1.4.2026 12-14, C124

Teymur Ismikhanov (TBA)

Wed 8.4.2026 12-14, C124

Andrew Cropper: Logic-based machine learning

Abstract: Inductive logic programming is a form of machine learning based on computational logic. It induces logical theories from data. I shall give a high-level overview to the field and its numerous applications, such as in drug discovery, game playing, and visual reasoning.

Wed 15.4.2026 12-14, C124

Phokion Kolaitis: TBA

Wed 22.4.2026 12-14, C124

Instead of the seminar there is a Workshop on Team Semantics in the city centre 20.-22.4.

Wed 29.4.2026 12-14, C124

Jarkko Kari: TBA

Wed 6.5.2026 12-14, C124

Lauri Hella & Kerkko Luosto: TBA

Past talks