14:00 – 14:25: Fenrong Liu, Reasons to believe in a social environment
14:25 – 14:50: Jan van Eijck, PDL Programs from Action Models
14:50 – 15:15: Olivier Roy, Collective agency, de dicto intentions, and pooled information
15:15 – 15:35: Break
15:35 – 16:00: Nina Gierasimczuk, Learning to Act: Qualitative Learning of Action Models
16:00 – 16:25: Frank Veltman, Hidden expectations
16:25 – 16:45: Break
16:45 – 17:10: Alexandru Baltag, Group Knowledge in Interrogative Epistemology
17:10 – 17:35: Johan van Benthem, Sabotage Logic Revisited
17:35 – 18:00: Davide Grossi & Zoé Christoff, From DeGroot processes to liquid democracy
Alexandru Baltag, Group Knowledge in Interrogative Epistemology
We use a version of the epistemic issue models of van Benthem and Minica to formalize an approach to knowledge that we call Interrogative Epistemology, in the spirit of Hintikka’s “interrogative model” of knowledge. According to our approach, an agent’s knowledge is shaped and limited by her interrogative agenda (as defined by her fundamental questions or “epistemic issues”). The dynamic correlate of this postulate is our Selective Learning principle: the agent’s agenda limits her potential for knowledge-acquisition. Only meaningful information, that is relevant to one’s issues, can really be learnt. This principle leads to a new semantics for public announcements, more restricted than the standard one. We use this approach to propose a new perspective on group knowledge, understood in terms of the epistemic potential of a group of agents: the knowledge that the group may come to possess in common (and thus act upon in a coordinated manner) after all members share their individual information. We argue that the standard notions of group knowledge studied in the literature, ranging from distributed knowledge to common knowledge, do not give us a good measure of a group’s epistemic potential. Common knowledge is too weak and too “static”, focusing on what the agents can coordinate upon only based on their actual, current knowledge (without any intra-group communication), thus disregarding testimonial knowledge. In contrast, distributed knowledge is too strong, being based on the assumption that agents can completely internalize all the testimonial evidence received from others, irrespective of the limitations posed by their own interrogative agendas. We show that a group’s true epistemic potential typically lies in between common knowledge and distributed knowledge. We propose a logical formalization of these concepts, which comes together with a complete axiomatization, and we use this setting to explain both the triumphs and the failures of collective knowledge, treating examples that range from “collective scientific knowledge” to the so-called “curse of the committee”. This is joint work with Rachel Boddy and Sonja Smets.
Johan van Benthem, Sabotage Logic Revisited
Sabotage modal logic was proposed in 2003 as a format for analyzing games that modify the graphs they are played on. We investigate some model-theoretic and proof-theoretic aspects of sabotage modal logic, which has largely come to be viewed as an early dynamic logic of graph change. Our first contribution is a characterization theorem for sabotage modal logic as a fragment of first-order logic which is invariant with respect to a natural notion of “sabotage bisimulation”. Our second contribution is a sound and complete tableau method for analyzing reasoning in sabotage modal logic. Finally, we identify and briefly explore a number of open research problems concerning sabotage modal logic that illuminate its complexity, placing it within the current landscape of modal logics that analyze model update, and, returning to the original motivation of sabotage, fixed-point logics for network games. This is joint work with Guillaume Aucher and Davide Grossi.
Jan van Eijck, PDL Programs from Action Models
We present an algorithm for constructing PDL programs from action models, based on a well known algorithm for constructing regular expressions from finite automata. The algorithm unifies the algebraic method of Brzozowski with the state removal method from many textbooks, and generalizes it to the case where the states have PDL preconditions.
The algorithm sheds new light on the reduction rules for the logic of communication and change (LCC). Another application is a structural characterization of the notion of action emulation (equivalence of action models).
Nina Gierasimczuk, Learning to Act: Qualitative Learning of Action Models
In this talk I will present a study of learning action models in dynamic epistemic logic. I will introduce a framework for actions seen as sets of transitions between propositional states and I will relate them to their DEL representations as action models. I will introduce and discuss various properties of actions and action models, e.g., new definition of determinism and (non-)conditionality. I will recall the basic learnability criteria: finite identifiability (conclusively inferring the appropriate action model in finite time) and identifiability in the limit (inconclusive convergence to the right action model). I will show that deterministic actions are finitely identifiable, while arbitrary (non-deterministic) actions require more learning power – they are identifiable in the limit. I will then move on to a particular learning method, i.e., learning via update, which proceeds via restriction of a space of events within a learning-specific action model. I will show how this method can be adapted to learn conditional and non-conditional action models. This is joint work with Thomas Bolander.
Davide Grossi & Zoé Christoff, From DeGroot processes to liquid democracy
We study a special case of opinion diffusion in networks where each agent holds binary opinions and follows a unique influencer. These simple “follow your guru” processes, which we name “Boolean DeGroot processes”, lie at the intersection of two more general frameworks for opinion diffusion: the stochastic model proposed by DeGroot, and the more recent approach, stemming from the literature on judgment aggregation, of “Propositional Opinion Diffusion”. We establish conditions for convergence of opinions in Boolean DeGroot processes and in a simple generalization of them, we show how these conditions can be captured by modal fixpoint logics, and we apply the convergence results to gain a novel insight into the collective decision-making system known as “liquid democracy”.
Fenrong Liu, Reasons to believe in a social environment
We present a logic which supports reasoning about an agent’s belief formation and belief change due to evidences provided by other agents in the society. We call this logic DEL-ES which stands for “Dynamic Epistemic Logic of Evidence Sources”. The term ‘evidence source’ refers to an agent in the society who provides an evidence to believe something to another agent. According to DEL-ES, if an agent has gathered a sufﬁcient number of evidences in support a given fact ϕ then, as a consequence, she should start to believe that ϕ is true. A sound and complete axiomatization for DEL-ES is given. We discuss some of its interesting properties and illustrate it in a concrete example. This is an ongoing work with Emiliano Lorini.
Olivier Roy, Collective agency, de dicto intentions, and pooled information
Large and unstructured groups have received little attention in the philosophical literature on collective agency. Most accounts focus either on small groups with common knowledge or on groups with a well-defined decision or aggregation mechanism. We propose an alternative account of collective agency for loose, unstructured groups. The key ideas are that collective agency is possible with de dicto plural intentions and information being sufficiently pooled. In transient groups, this type of agency may only emerge for short periods of time. This account generalizes a number of existing theories of collective agency while still allowing to distinguish loose collective agency from aggregate behavior or movements. This is joint work with Anne Schwenkenbecher (Murdoch U, Perth).
Frank Veltman, Hidden expectations
In this talk I will have to say something about a variety of topics:
- epistemic modalities (What is the difference between `may’ and `might’?)
- conditionals (There is a third kind of conditionals, “somewhere between” indicatives and counterfactuals.)
- gradable adjectives (Why is it is normal to be neither tall nor short?)
- generic sentences (Why accept `lions have manes’ rather than `lions don’t have manes’? How come `Boys don’t cry’ seems to express a norm rather than a fact?).
What these topics have in common is that one cannot explain the meaning — not even the logical properties — of the expressions involved without explaining how they affect people’s “expectations”. This can best be done in a framework in which the meaning of a sentence is not equated with its truth conditions but with its (potential) impact on the intentional state of an addressee.