The introduction of social robots into society will require that they follow ethical principles which go beyond consequentialism.

In this paper, I show how to apply the principle of double effect to solve an ethical dilemma involving robots studied by Alan Winfield and colleagues.

The principle of double effect states conditions for ethically acceptable behavior when there are both positive and negative consequences of an action.

I propose a formal semantics with actions, causes, intentions, and utilities based upon the work of Judea Pearl, John Horty and others. With this formal semantics,

the question of whether an action is permitted according to the principle of double effect is reduced to deciding whether a certain formula is true or otherwise.

When: October 12 at 10:30

Where: Building 303, room 026 – Technical University of Denmark (DTU)

Host: Associate Professor Thomas Bolander

]]>

A well-studied phenomenon in network theory since the 1970s are optimal schedules to distribute information by one-to-one communication between nodes. One can take these communicative actions to be telephone calls, and protocols to spread information this way are known as gossip protocols or epidemic protocols. Statistical approaches to gossip have taken a large flight since then, witness for example the survey "Epidemic Information Dissemination in Distributed Systems" by Eugster et al. (IEEE Computer, 2004). It is typical to assume a global scheduler who executes a possibly non-deterministic or randomized protocol. A departure from this methodology is to investigate epistemic gossip protocols, where an agent (node) will call another agent not because it is so instructed by a scheduler, but based on its knowledge or ignorance of the distribution of secrets over the network and of other agents' knowledge or ignorance of that. Such protocols are distributed and do not need a central scheduler. This comes at a cost: they may take longer to terminate than non-epistemic, globally scheduled, protocols. A number of works have appeared over the past years (Apt et al., Attamah et al., van Ditmarsch et al., van Eijck & Gatting, Herzig & Maffre) of which we present a survey, including open problems yet to be solved by the community.

Everyone is welcome

]]>

There are rich traditions in cognitive science psychology, linguistics, philosophy, artificial intelligence and logic centring on the theme of perspective-taking: the ability of agents to view information from different viewpoints, and to take the viewpoint of other agents into account. Expressions such as Theory of Mind, Conversational Implicature, Social Cognition, and the Semantics of Propositional Attitudes indicate some relevant themes.

The

For more details:

http://cadillac-dk.weebly.com/2016-cognitive-linguistic-and-logical-aspects-of-perspective-taking.html

]]>

https://cadillac.compute.dtu.dk

The CADILLAC workshop series will continue the Modality and Modalities workshop series (http://modalityandmodalities.weebly.com), but will extend its scope beyond modal logic, to cover applications of logical methods in philosophy, linguistics and cognitive science. The CADILLAC workshops will be organised by the CADILLAC network (http://cadillac-dk.weebly.com). This first edition is organised by Martin Mose Bentzen, Nina Gierasimczuk and Thomas Bolander.

It is the first scientific conference to take place in Christiania, and it will feature a Science and Cocktails talk (http://www.scienceandcocktails.org) by Rineke Verbrugge. The third day of the conference is a special theme day on

We wish to continue the tradition from the Modality and Modalities workshops, meaning that one of the goals of the workshop is to serve as a gathering of the local researchers in the CADILLAC areas: Dynamics, Interaction, Logic, Language and Computation. Everybody is very welcome to attend the workshop, and it is free of charge. Please consult the workshop web page for details on how to register.

https://cadillac.compute.dtu.dk

]]>

Modalities, conditionals, and values:

A symposium on philosophical logic in celebration of the centenary of

Georg Henrik von Wright.

http://www.cs.helsinki.fi/u/hakli/vw/

The meeting will be held in Helsinki on May 23-25, 2016.

Confirmed international participants include

Patrick Blackburn

Roy Dyckhoff

Andreas Herzig

Valentin Goranko

Melvin Fitting

Nicola Olivetti

Eric Pacuit

Gabriella Pigozzi

Giovanni Sartor

Extended abstracts (2-3 pages) of contributed papers should be

submitted through EasyChair:

https://easychair.org/conferences/?conf=vonwright2016

Contributed papers will be selected through a peer review process on

the basis of scientific quality as well as of thematic relevance for

the symposium (see a detailed thematic description below).

Important dates:

February 7, 2016: Deadline of submission

March 1, 2016: Notification

Organizers:

Sara Negri

http://www.helsinki.fi/~negri/

http://www.helsinki.fi/teoreettinenfilosofia/henkilosto/Negri.htm

Raul Hakli

Conference contact:

vonwright-symposium@helsinki.fi

******************************************************************

Georg Henrik von Wright was one of the early pioneers of philosophical

logic, with his monograph ``An Essay in Modal Logic'' of 1951, and his

invention of a whole new field, namely deontic logic.

Recent developments in the study of modal logic and related

philosophical logics in Helsinki go hand-in-hand with a revival of

philosophical logic at the international level, with new ideas,

methods and fields of application stemming from the pioneering work

that started in the 1950s with the modern development in the

philosophical investigation of modalities.

In modal logic, syntactic and semantic methods obtain a fruitful blend

through correspondence theory and their blend has in recent years

created a new momentum in the development of uniform and comprehensive

inferential systems, something which was considered an impossibility

until recently. In the theory of conditionals, the logical concept of

implication or consequence is set in a much broader light. Unlike

material implication, conditionals are not truth functionals. They

also escape an interpretation in terms of alethic modality, thus

creating a challenge to logicians. A number of scholars have faced the

problem of interpreting conditionals and creating good inferential

systems for them.

In addition to modal, deontic, conditional, and tense logic, the

symposium will focus on the far-reaching developments of other of von

Wright's seminal contributions, such as the logic of preference, with

implications in areas such as economics, game theory, social choice

theory, and artificial intelligence, as well as pervasive themes of

von Wright's production, such as reflection on the role of logic in

philosophy.

The workshop will bring together some established experts in the field

as well as some younger scholars. Contributed papers are welcome in

the research area in logic stemming from von Wright's investigations,

including the following:

- modal logic

- tense logic

- conditional logic

- deontic logic

- logic of preference

- logic of action

- generalized semantics (neighbourhood semantics, game semantics)

]]>

13.00-14.00

14.15-15.15

Iris van de Pol

In this talk I will discuss the complexity of theory of mind, which is the cognitive phenomenon that people attribute mental states to each other, like beliefs, desires, and intentions, and use these to predict or explain each other’s behavior. I present a computational-level model of theory of mind—based on dynamic epistemic logic—and analyze its (parameterized) complexity. In particular I show that the model is PSPACE-complete and that the modal depth of a formula does not influence its complexity.

This talk is based on my master’s thesis. For a pdf, see: http://goo.gl/21FG2G.

--

Jakub Szymanik.

Inspired by the logical analysis of backward induction and cognitive science experiments, we investigate the computational complexity of the reasoning in extensive form dynamic games. We formalize the computational complexity of a general decision problem the players face, assuming common knowledge of rationality. We show that such defined problem is PTIME-complete. Moreover, we provide a more refined analysis of the complexity of particular game trials which takes into account alternation type of the game and pay-off distribution. We conclude with some mathematical and cognitive open problems.

--

Please sign up for the talk at latest on 13 January, so I know how much beer and snacks to buy.

All are welcome! (Please distribute this email to any others you think might be interested).

Best regards,

Thomas Bolander]]>

Traditionally, formulas are written on a single line. S. Curtis and G. Lowe suggested a more visually appealing alternative for the case of binary relations: using graphs for expressing properties and reasoning about relations in a natural way. We extended their approach to diagrams that are sets of graphs.

More specifically, in this setting, diagrams corresponds to sentences and transformations on graphs diagrams correspond to inference rules, that can be used to infer a diagram from a set of diagrams taken as hypothesis. The basic intuitions are quite simple, leading to playful and powerful systems. Our system fits very well as a tool for perform inferences on algebras of binary relations, but we developed it to treat positive, negative, and intuitionistic information, ranging from the geometric fragment to the whole of first order logic on binary relations.

In this talk I summarize these achievements, presenting some on going work on extending the previous systems to deal with the equations and containments of the distributive, division, and power allegories, being tabular or not. Given the connection between classical modal logics and algebras of relations developed by E. Orlowska and her collaborators, our graphical system may be also explored as a tool for perform inferences in modal logic.]]>

- Nina Gierasimczuk (http://www.ninagierasimczuk.com) is employed as assistant professor from 1 January 2015.

- Antti Kuusisto (http://www.sis.uta.fi/~ak75987/) is employed as postdoc from 1 September 2015 to 31 August 2017. Antti is funded by the (highly competitive) HC Ørsted Postdoc programme at DTU, co-funded by Marie Curie Actions.

We look very much forward to welcome these two excellent people to the CADILLAC/Copenhagen logic community. Please keep them in mind for your future logic-related activities in "the greater Copenhagen area" (and in general!).

Best,

Thomas Bolander

]]>

Two talks by Gert Jan Hofstede in Copenhagen, 7 April

--

Gert Jan Hofstede works on modelling social behaviour, and is famous both for his work on social agents and his work on dimensions of culture together with his father, Geert Hofstede (

—

I will argue that logic in the customary sense is not the essential ingredient for social intelligence. We people use status-power thinking in a group-based relational world as the core mechanism driving our actions. If agents do not understand the status-power world they will be condemned to remain ‘idiot savants’.

How can agents acquire this kind of reasoning? We need 1) good, relevant theories from social science and/or biology, 2) an ontology of mid-range constructs that link these theories to the agent world, and 3) instantiations of constructs in the agent world. I’ll provide some examples.

Links:

·

·

·

·

—

Proof Assistants and Related Tools - The PART Project

Talks:

12:15-12:45 Jørgen Villadsen (DTU Compute): Computer-Checked Meta-Logic

13:00-13:30 Jesper Bengtson (ITU): Software Verification Using Proof Assistants

14:00-14:30 Sebastian Alexander Mödersheim (DTU Compute): Integrating Automated and Interactive Protocol Verification (joint work with Achim Brucker, SAP Research)

14:45-15:15 Ximeng Li (DTU Compute): Coq for Programming Language Proofs - A Personal Experience

Please find the full program and abstracts here:

]]>