This special issue of the Logic Journal of the IGPL includes revised and updated versions of the best work presented at the fourth edition of the workshop Formal Ap- proaches to Multi-Agent Systems, FAMAS'09, which took place in Turin, Italy, from 7 to 11 September, 2009, under the umbrella of the Multi-Agent Logics, Languages, and Organisations Federated Workshops (MALLOW). -/- Just like its predecessor, research reported in this FAMAS 2009 special issue is very much inspired by (...) practical concerns. This time the authors of all the five selected papers are concerned with knowledge and beliefs in multi-agent settings: How to create a group belief in a fair way from individual plausibility orderings? How to close gaps and resolve ambiguities in a tractable way, when information comes from multiple sources? How to reason about a spatial environment? How to compare the strengths of an agent's beliefs in a principled way? How to decide as efficiently as possible whether a given formula concerning group beliefs is valid? These questions and their answers lead to a multi-faceted and at the same time coherent special issue. We concisely introduce the five articles. (shrink)

Over the last decade, multi-agent systems have come to form one of the key tech- nologies for software development. The Formal Approaches to Multi-Agent Systems (FAMAS) workshop series brings together researchers from the fields of logic, theoreti- cal computer science and multi-agent systems in order to discuss formal techniques for specifying and verifying multi-agent systems. FAMAS addresses the issues of logics for multi-agent systems, formal methods for verification, for example model (...) check- ing, and formal approaches to cooperation, multi-agent planning, communication, coordination, negotiation, games, and reasoning under uncertainty in a distributed environment. In 2007, the third FAMAS workshop, FAMAS'007, was one of the agent workshops gathered together under the umbrella of Multi-Agent Logics, Languages, and Organ- isations - Federated Workshops, MALLOW'007, taking place from 3 to 7 September 2007 in Durham. This current special issue of the Logic Journal of the IGPL gathers together the revised and updated versions of the five best FAMAS'007 contributions. (shrink)

We formally introduce a novel, yet ubiquitous, category of norms: norms of instrumentality. Norms of this category describe which actions are obligatory, or prohibited, as instruments for certain purposes. We propose the Logic of Agency and Norms (LAN) that enables reasoning about actions, instrumentality, and normative principles in a multi-agent setting. Leveraging LAN , we formalize norms of instrumentality and compare them to two prevalent norm categories: norms to be and norms to do. Last, we pose principles relating (...) the three categories and evaluate their validity vis-à-vis notions of deliberative acting. On a technical note, the logic will be shown decidable via the finite model property. (shrink)

Although virtue ethics has repeatedly been proposed as a suitable framework for the development of artificial moral agents, it has been proven difficult to approach from a computational perspective. In this work, we present the first technical implementation of artificial virtuous agents in moral simulations. First, we review previous conceptual and technical work in artificial virtue ethics and describe a functionalistic path to AVAs based on dispositional virtues, bottom-up learning, and top-down eudaimonic reward. We then provide the details (...) of a technical implementation in a moral simulation based on a tragedy of the commons scenario. The experimental results show how the AVAs learn to tackle cooperation problems while exhibiting core features of their theoretical counterpart, including moral character, dispositional virtues, learning from experience, and the pursuit of eudaimonia. Ultimately, we argue that virtue ethics provides a compelling path toward morally excellent machines and that our work provides an important starting point for such endeavors. (shrink)

Using a knowledge-based approach, we derive a protocol, MACOM1, for the sequence transmission problem from one agent to a group of agents. The protocol is correct for communication media where deletion and reordering errors may occur. Furthermore, it is shown that after k rounds the agents in the group attain depth k general knowledge about the members of the group and the values of the messages. Then, we adjust this algorithm for multi-agent communication for the (...) process of teamwork. MACOM1 solves the sequence transmission problem from one agent to a group of agents, but does not fully comply with the type of dialogue communication needed for teamwork. The number of messages being communicated per communication between the initiator and the other agents from the group can differ. Furthermore, the teamwork process can require the communication algorithm to handle changes of initiator. We show the adjustments that have to be made to MACOM1 to handle these properties of teamwork. For the new multi-agent communication algorithm, MACOM2, it is shown that the gaining of knowledge required for a successful teamwork process is still guaranteed. (shrink)

The purpose of this study is to compile a selection of the various formalisms found in conversation theory to introduce readers to Pask's discursive algebra. In this way, the text demonstrates how concept sharing and concept formation by means of the interaction of two participants may be formalized. The approach taken in this study is to examine the formal notation system used by Pask and demonstrate how such formalisms may be used to represent concept sharing and concept formation through conversation. (...) The compilation of the discursive algebra using the framework provided by conversation theory could potentially be used as an auxiliary framework to study conversational interactions in multi-agent systems theory. (shrink)

In this paper we analyze how the semantics of the Alternating-time Temporal Logic ATL$^*$ deals with agents' commitments to strategies in the process of formula evaluation. In (\acro{atl}$^*$), one can express statements about the strategic ability of an agent (or a coalition of agents) to achieve a goal $\phi$ such as: ``agent $i$ can choose a strategy such that, if $i$ follows this strategy then, no matter what other agents do, $\phi$ will always be true''. (...) However, strategies in \acro{atl} are \emph{revocable} in the sense that in the evaluation of the goal $\phi$ the agent $i$ is no longer restricted by the strategy she has chosen in order to reach the state where the goal is evaluated. -/- Here we discuss some alternatives leading to amendments of that semantics. In particular, we consider variants of \acro{atl}$^*$ where strategies, on the contrary, are \emph{irrevocable}. Unlike in the standard semantics of \acro{atl}, memory plays an essential role in the semantics based on irrevocable strategies. -/- Further, we propose and discuss various syntactic and semantics mechanisms for handling commitments to strategies and release from such commitments in the semantics of ATL$^*$, leading to more expressive and semantically refined versions of that logic. (shrink)

Inspired by recent progress in multi-agent Reinforcement Learning (RL), in this work we examine the collective intelligent behaviour of theoretical universal agents by introducing a weighted mixture operation. Given a weighted set of agents, their weighted mixture is a new agent whose expected total reward in any environment is the corresponding weighted average of the original agents' expected total rewards in that environment. Thus, if RL agent intelligence is quantified in terms of performance (...) across environments, the weighted mixture's intelligence is the weighted average of the original agents' intelligences. This operation enables various interesting new theorems that shed light on the geometry of RL agent intelligence, namely: results about symmetries, convex agent-sets, and local extrema. We also show that any RL agent intelligence measure based on average performance across environments, subject to certain weak technical conditions, is identical (up to a constant factor) to performance within a single environment dependent on said intelligence measure. (shrink)

The emergence of Large Language Models (LLMs) and advancements in Artificial Intelligence (AI) offer an opportunity for computational social science research at scale. Building upon prior explorations of LLM agent design, our work introduces a simulated agent society where complex social relationships dynamically form and evolve over time. Agents are imbued with psychological drives and placed in a sandbox survival environment. We conduct an evaluation of the agent society through the lens of Thomas Hobbes's seminal Social (...) Contract Theory (SCT). We analyze whether, as the theory postulates, agents seek to escape a brutish "state of nature" by surrendering rights to an absolute sovereign in exchange for order and security. Our experiments unveil an alignment: Initially, agents engage in unrestrained conflict, mirroring Hobbes's depiction of the state of nature. However, as the simulation progresses, social contracts emerge, leading to the authorization of an absolute sovereign and the establishment of a peaceful commonwealth founded on mutual cooperation. This congruence between our LLM agent society's evolutionary trajectory and Hobbes's theoretical account indicates LLMs' capability to model intricate social dynamics and potentially replicate forces that shape human societies. By enabling such insights into group behavior and emergent societal phenomena, LLM-driven multi-agent simulations, while unable to simulate all the nuances of human behavior, may hold potential for advancing our understanding of social structures, group dynamics, and complex human systems. (shrink)

The aim of this paper is to discuss the “Framework for M&S with Agents” (FMSA) proposed by Zeigler et al. [2000, 2009] in regard to the diverse epistemological aims of agent simulations in social sciences. We first show that there surely are great similitudes, hence that the aim to emulate a universal “automated modeler agent” opens new ways of interactions between these two domains of M&S with agents. E.g., it can be shown that the multi-level (...) conception at the core of the FMSA is similar in both contexts: notions of “levels of system specifi cation”, “behavior of models”, “simulator”and “endomorphic agents” can be partially translated in the terms linked to the “denotational hierarchy” (DH) recently introduced in a multi-level centered epistemology of M&S. Second, we suggest considering the question of “credibility” of agent M&S in social sciences when we do not try to emulate but only to simulate target systems. Whereas a stringent and standardized treatment of the heterogeneous internal relations (in the DH) between systems of formalisms is the key problem and the essential challenge in the scope of Agent M&S driven engineering, it is urgent too to address the problem of the external relations (and of the external validity, hence of the epistemic power and credibility) of such levels of formalisms in the specific domains of agent M&S in social sciences, especially when we intend to introduce the concepts of activity tracking. (shrink)

We develop a conceptually clear, intuitive, and feasible decision procedure for testing satisfiability in the full multi\-agent epistemic logic \CMAELCD\ with operators for common and distributed knowledge for all coalitions of agents mentioned in the language. To that end, we introduce Hintikka structures for \CMAELCD\ and prove that satisfiability in such structures is equivalent to satisfiability in standard models. Using that result, we design an incremental tableau-building procedure that eventually constructs a satisfying Hintikka structure for every satisfiable (...) input set of formulae of \CMAELCD\ and closes for every unsatisfiable input set of formulae. (shrink)

Whether collective agency is a coherent concept depends on the theory of agency that we choose to adopt. We argue that the enactive theory of agency developed by Barandiaran, Di Paolo and Rohde (2009) provides a principled way of grounding agency in biological organisms. However the importance of biological embodiment for the enactive approach might lead one to be skeptical as to whether artificial systems or collectives of individuals could instantiate genuine agency. To explore this issue we contrast the concept (...) of collective agency with multi-agent systems and multi-system agents, and argue that genuinely collective agents instantiate agency at both the collective level and at the level of the component parts. Developing the enactive model, we propose understanding agency – both at the level of the individual and of the collective – as spectra that are constituted by dimensions that vary across time. Finally, we consider whether collectives that are not merely metaphorically ‘agents’ but rather are genuinely agentive also instantiate subjectivity at the collective level. We propose that investigations using the perceptual crossing paradigm suggest that a shared lived perspective can indeed emerge but this should not be conflated with a collective first-person perspective, for which material integration in a living body may be required. (shrink)

This work provides proof-search algorithms and automated counter-model extraction for a class of STIT logics. With this, we answer an open problem concerning syntactic decision procedures and cut-free calculi for STIT logics. A new class of cut-free complete labelled sequent calculi G3LdmL^m_n, for multi-agent STIT with at most n-many choices, is introduced. We refine the calculi G3LdmL^m_n through the use of propagation rules and demonstrate the admissibility of their structural rules, resulting in auxiliary calculi Ldm^m_nL. In the single- (...) class='Hi'>agent case, we show that the refined calculi Ldm^m_nL derive theorems within a restricted class of (forestlike) sequents, allowing us to provide proof-search algorithms that decide single-agent STIT logics. We prove that the proof-search algorithms are correct and terminate. (shrink)

Counterfactual attitudes like imagining, dreaming, and wishing create a problem for the standard formal semantic theory of de re attitude ascriptions. I show how the problem can be avoided if we represent an agent's attitudinal possibilities using "multi-centered worlds", possible worlds with multiple distinguished individuals, each of which represents an individual with whom the agent is acquainted. I then present a compositional semantics for de re ascriptions according to which singular terms are "assignment-sensitive" expressions and attitude verbs (...) are "assignment shifters". (shrink)

We present a multi-agent system where agents can cooperate to solve a system of dependent tasks, with agents having the capability to explore a solution space, make inferences, as well as query for information under a limited budget. Re-exploration of the solution space takes place by an agent when an older solution expires and is thus able to adapt to dynamic changes in the environment. We investigate the effects of task dependencies, with highly-dependent graph G_40 (...) (a well-known program graph that contains 40 highly interlinked nodes, each representing a task) and less-dependent graphs G_18 (a program graph that contains 18 tasks with fewer links), increasing the speed of the agents and the complexity of the problem space and the query budgets available to agents. Specifically, we evaluate trade-offs between the agent's speed and query budget. During the experiments, we observed that increasing the speed of a single agent improves the system performance to a certain point only, and increasing the number of faster agents may not improve the system performance due to task dependencies. Favoring faster agents during budget allocation enhances the system performance, in line with the "Matthew effect." We also observe that allocating more budget to a faster agent gives better performance for a less-dependent system, but increasing the number of faster agents gives a better performance for a highly-dependent system. (shrink)

Existing studies of the multi-group dynamics of prejudiced societies focus on the social- psychological knowledge behind the relevant processes. We instead create a multi-agent framework that simulates the propagation of prejudice and measures its tangible impact on prosperity. Levels of prosperity are tracked for individuals as well as larger social structures including groups and factions. We model social interactions using the Continuous Prisoner's Dilemma (CPD) and a new agent type called a prejudiced agent. Our simulations (...) show that even modeling prejudice as an exclusively out-group phenomenon nonetheless generates implicit in-group promotion, leading to higher relative prosperity of a prejudiced population. This skew in prosperity is seen to be correlated to factors such as size difference between groups, and the fraction of prejudiced agents in a group. Our model is a step towards a deeper understanding of the origins, propagation, and ramifications of prejudice through simulative studies grounded in apt theoretical backgrounds. (shrink)

In this work we answer a long standing request for temporal embeddings of deontic STIT logics by introducing the multi-agent STIT logic TDS . The logic is based upon atemporal utilitarian STIT logic. Yet, the logic presented here will be neutral: instead of committing ourselves to utilitarian theories, we prove the logic TDS sound and complete with respect to relational frames not employing any utilitarian function. We demonstrate how these neutral frames can be transformed into utilitarian temporal frames, (...) while preserving validity. Last, we discuss problems that arise from employing binary utility functions in a temporal setting. (shrink)

Many collective decision making problems have a combinatorial structure: the agents involved must decide on multiple issues and their preferences over one issue may depend on the choices adopted for some of the others. Voting is an attractive method for making collective decisions, but conducting a multi-issue election is challenging. On the one hand, requiring agents to vote by expressing their preferences over all combinations of issues is computationally infeasible; on the other, decomposing the problem into several (...) elections on smaller sets of issues can lead to paradoxical outcomes. Any pragmatic method for running a multi-issue election will have to balance these two concerns. We identify and analyse the problem of generating an agenda for a given election, specifying which issues to vote on together in local elections and in which order to schedule those local elections. (shrink)

Research in multi-agent systems typically assumes a regulative model of social practice. This model starts with agents who are already capable of acting autonomously to further their individual ends. A social practice, according to this view, is a way of achieving coordination between multiple agents by restricting the set of actions available. For example, in a world containing cars but no driving regulations, agents are free to drive on either side of the road. To prevent (...) collisions, we introduce driving regulations, insisting that everyone drives on the left hand side of the road. We accept this limitation on our freedom because it lowers the probability of a collision. -/- This paper describes AI systems that are based on the constitutive view of social practice. According to this alternative view, there are certain actions that are only available to the agent because he is participating in a practice of a certain sort. For example, you can swing a peculiarly shaped piece of wood without participating in any particular practice - but this action will only constitute a strike if you are participating in a game of baseball. I can raise my hand whenever I like, but this only counts as voting for the motion within the institution of voting. But not all the examples are from frivolous, optional or institutional activities. Sellars and Brandom argue that even the fundamental act of assertion is only possible within the context of a particular social practice: the Game of Giving and Asking for Reasons. -/- The constitutive view, which is familiar to philosophers, is relatively unknown within the AI community. -/- The central aim of this paper is to show how the constitutive view of practice can be put to use in AI applications. (shrink)

We define a formal semantics of conditionals based on _normatively ideal worlds_. Such worlds are described informally by Armgardt (Gabbay D, Magnani L, Park W, Pietarinen A-V (eds) Natural arguments: a tribute to john woods, College Publications, London, pp 699–708, 2018) to address well-known problems of the counterfactual approach to causation. Drawing on Armgardt’s proposal, we use iterated conditionals in order to analyse causal relations in scenarios of multi-agent interaction. This results in a refined counterfactual approach to causal (...) responsibility in legal contexts, which solves overdetermination problems in an intuitively accessible manner. (shrink)

Axiom weakening is a novel technique that allows for fine-grained repair of inconsistent ontologies. In a multi-agent setting, integrating ontologies corresponding to multiple agents may lead to inconsistencies. Such inconsistencies can be resolved after the integrated ontology has been built, or their generation can be prevented during ontology generation. We implement and compare these two approaches. First, we study how to repair an inconsistent ontology resulting from a voting-based aggregation of views of heterogeneous agents. Second, we (...) prevent the generation of inconsistencies by letting the agents engage in a turn-based rational protocol about the axioms to be added to the integrated ontology. We instantiate the two approaches using real-world ontologies and compare them by measuring the levels of satisfaction of the agents w.r.t. the ontology obtained by the two procedures. (shrink)

The so-called Baldwin Effect generally says how learning, as a form of ontogenetic adaptation, can influence the process of phylogenetic adaptation, or evolution. This idea has also been taken into computation in which evolution and learning are used as computational metaphors, including evolving neural networks. This paper presents a technique called evolving self-taught neural networks – neural networks that can teach themselves without external supervision or reward. The self-taught neural network is intrinsically motivated. Moreover, the self-taught neural network is the (...) product of the interplay between evolution and learning. We simulate a multi-agent system in which neural networks are used to control autonomous agents. These agents have to forage for resources and compete for their own survival. Experimental results show that the interaction between evolution and the ability to teach oneself in self-taught neural networks outperform evolution and self-teaching alone. More specifically, the emergence of an intelligent foraging strategy is also demonstrated through that interaction. Indications for future work on evolving neural networks are also presented. (shrink)

Making good decisions depends on having accurate information – quickly, and in a form in which it can be readily communicated and acted upon. Two features of medical practice can help: deliberation in groups and the use of scores and grades in evaluation. We study the contributions of these features using a multi-agent computer simulation of groups of physicians. One might expect individual differences in members’ grading standards to reduce the capacity of the group to discover the facts (...) on which well-informed decisions depend. Observations of the simulated groups suggest on the contrary that this kind of diversity can in fact be conducive to epistemic performance. Sometimes, it is adopting common standards that may be expected to result in poor decisions. (shrink)

Causal reasoning and game-theoretic reasoning are fundamental topics in artificial intelligence, among many other disciplines: this paper is concerned with their intersection. Despite their importance, a formal framework that supports both these forms of reasoning has, until now, been lacking. We offer a solution in the form of (structural) causal games, which can be seen as extending Pearl's causal hierarchy to the game-theoretic domain, or as extending Koller and Milch's multi-agent influence diagrams to the causal domain. We then (...) consider three key questions: i) How can the (causal) dependencies in games - either between variables, or between strategies - be modelled in a uniform, principled manner? ii) How may causal queries be computed in causal games, and what assumptions does this require? iii) How do causal games compare to existing formalisms? To address question i), we introduce mechanised games, which encode dependencies between agents' decision rules and the distributions governing the game. In response to question ii), we present definitions of predictions, interventions, and counterfactuals, and discuss the assumptions required for each. Regarding question iii), we describe correspondences between causal games and other formalisms, and explain how causal games can be used to answer queries that other causal or game-theoretic models do not support. Finally, we highlight possible applications of causal games, aided by an extensive open-source Python library. (shrink)

We apply spatialized game theory and multi-agent computational modeling as philosophical tools: (1) for assessing the primary social psychological hypothesis regarding prejudice reduction, and (2) for pursuing a deeper understanding of the basic mechanisms of prejudice reduction.

This paper presents a human-computer interaction model with a three layers learning mechanism in a pervasive environment. We begin with a discussion around a number of important issues related to human-computer interaction followed by a description of the architecture for a multi-agent cooperative design system for pervasive computing environment. We present our proposed three- layer HCI model and introduce the group formation algorithm, which is predicated on a dynamic sharing niche technology. Finally, we explore the cooperative reinforcement learning (...) and fusion algorithms; the paper closes with concluding observations and a summary of the principal work and contributions of this paper. (shrink)

(See also the separate entry for the volume itself.) This introduction has three parts. The first providing an overview of some main lines of research in deontic logic: the emergence of SDL, Chisholm's paradox and the development of dyadic deontic logics, various other puzzles/challenges and areas of development, along with philosophical applications. The second part focus on some actual and potential fruitful interactions between deontic logic, computer science and artificial intelligence. These include applications of deontic logic to AI knowledge representation (...) in legal systems, to modelling computer systems where it is expected that sub-ideal states will emerge and require countermeasures, to norm-governed human interactions with computer systems, and to the representation of some features of multi-agent systems where different agent-like computer systems interact with one another. The third and final part briefly groups and previews the papers in the anthology. (shrink)

Multiple Authors - please see paper attached. -/- AI systems have seen dramatic advancement in recent years, bringing many applications that pervade our everyday life. However, we are still mostly seeing instances of narrow AI: many of these recent developments are typically focused on a very limited set of competencies and goals, e.g., image interpretation, natural language processing, classification, prediction, and many others. We argue that a better study of the mechanisms that allow humans to have these capabilities can help (...) us understand how to imbue AI systems with these competencies. We focus especially on D. Kahneman’s theory of thinking fast and slow , and we propose a multi-agent AI architecture (called SOFAI, for SlOw and Fast AI) where incoming problems are solved by either system 1 (or "fast") agents (also called "solvers"), that react by exploiting only past experience, or by system 2 (or "slow") agents, that are deliberately activated when there is the need to reason and search for optimal solutions beyond what is expected from the system 1 agent. (shrink)

I offer an account of the experience of acting that demonstrates how agentive aspects of experience associated with the execution of intentions are richly integrated with perceptual aspects associated with parts of action taking place in the publicly observable world. On the view I elucidate, the experience of acting is often both an engagement with the world and a type of intimate acquaintance with it. In conscious action the agent consciously intervenes in the world and consciously experiences the world (...) she is changing. In section one, I discuss extant accounts of the experience of acting, noting deficiencies. In sections two and three, I develop my own account, drawing on Casey O’Callaghan’s work on multi-modal perception. In the conclusion, I discuss ramifications for psychology and philosophy. (shrink)

[Müller, Vincent C. (ed.), (2016), Fundamental issues of artificial intelligence (Synthese Library, 377; Berlin: Springer). 570 pp.] -- This volume offers a look at the fundamental issues of present and future AI, especially from cognitive science, computer science, neuroscience and philosophy. This work examines the conditions for artificial intelligence, how these relate to the conditions for intelligence in humans and other natural agents, as well as ethical and societal problems that artificial intelligence raises or will raise. The key issues (...) this volume investigates include the relation of AI and cognitive science, ethics of AI and robotics, brain emulation and simulation, hybrid systems and cyborgs, intelligence and intelligence testing, interactive systems, multi-agent systems, and superintelligence. Based on the 2nd conference on “Theory and Philosophy of Artificial Intelligence” held in Oxford, the volume includes prominent researchers within the field from around the world. (shrink)

Scientific realists argue that a good track record of multi-agent, and multiple method, validation of empirical claims is itself evidence that those claims, at least partially and approximately, reflect ways nature actually is independent of the ways we conceptualize it. Constructivists contend that successes in validating empirical claims only suffice to establish that our ways of modelling the world, our “constructions,” are useful and adequate for beings like us. This essay presents a thought experiment in which beings like (...) us intersubjectively validate claims about properties of particular things in nature under conditions in which those beings have profoundly different personal phenomenological experiences of those properties. I submit that the thought experiment scenario parallels our actual situation, and argue that this shows that successes in intersubjectively validating empirical claims are indeed enough to claim victory for the realist. More specifically, I champion a variation of realism that marries Ronald Giere’s brand of ‘perspectival realism’ with Philip Kitcher’s ‘real realism,’ and posits that causal relations between ourselves and properties instantiated in nature ground our references to the relevant properties even though our conceptions of them are perspective relative. (shrink)

This article presents modal versions of resource-conscious logics. We concentrate on extensions of variants of linear logic with one minimal non-normal modality. In earlier work, where we investigated agency in multi-agent systems, we have shown that the results scale up to logics with multiple non-minimal modalities. Here, we start with the language of propositional intuitionistic linear logic without the additive disjunction, to which we add a modality. We provide an interpretation of this language on a class of Kripke (...) resource models extended with a neighbourhood function: modal Kripke resource models. We propose a Hilbert-style axiomatisation and a Gentzen-style sequent calculus. We show that the proof theories are sound and complete with respect to the class of modal Kripke resource models. We show that the sequent calculus admits cut elimination and that proof-search is in PSPACE. We then show how to extend the results when non-commutative connectives are added to the language. Finally, we put the l.. (shrink)

Over the past decade or so, a new interdisciplinary field has emerged in the ground between, on the one hand, computer science – and artificial intelligence in particular – and, on the other, the area of philosophy concentrating on the language and structure of argument. There are now hundreds of researchers worldwide who would consider themselves a part of this nascent community. Various terms have been proposed for the area, including "Computational Dialectics," "Argumentation Technology," and "Argument-based Computing," but the term (...) that has stuck is simply Argument & Computation. It encompasses several specific strands of research, such as: . the use of theories of argument, and dialectic in particular, in the design and implementation of protocols for multi-agent action and communication; . the application of theories of argument and rhetoric in natural language processing and affective computing; . the use of argument-based structures for autonomous reasoning in artificial intelligence, and in particular, for defeasible reasoning; . computer-supported collaborative argumentation – the implementation of software tools for enabling online argument in domains such as education and e-government. These strands come together to form the core of a research field that covers parts of artificial intelligence (AI), philosophy, linguistics, and cognitive science, but, increasingly, is building an identity of its own. (shrink)

The problem of personal identity is a longstanding philosophical topic albeit without final consensus. In this article the somewhat similar problem of AI identity is discussed, which has not gained much traction yet, although this investigation is increasingly relevant for different fields, such as ownership issues, personhood of AI, AI welfare, brain–machine interfaces, the distinction between singletons and multi-agent systems as well as to potentially support finding a solution to the problem of personal identity. The AI identity problem (...) analyses the criteria for two AIs to be considered the same at different points in time. Two approaches to tackle the problem are proposed: One is based on the personal identity problem and the concept of computational irreducibility, while the other one applies multi-factor authentication to the AI identity problem. Also, a range of scenarios is examined regarding AI identity, such as replication, fission, fusion, switch off, resurrection, change of hardware, transition from non-sentient to sentient, journey to the past, offspring and identity change. (shrink)

Standpoint logic is a recently proposed formalism in the context of knowledge integration, which advocates a multi-perspective approach permitting reasoning with a selection of diverse and possibly conflicting standpoints rather than forcing their unification. In this paper, we introduce nested sequent calculi for propositional standpoint logics---proof systems that manipulate trees whose nodes are multisets of formulae---and show how to automate standpoint reasoning by means of non-deterministic proof-search algorithms. To obtain worst-case complexity-optimal proof-search, we introduce a novel technique in the (...) context of nested sequents, referred to as "coloring," which consists of taking a formula as input, guessing a certain coloring of its subformulae, and then running proof-search in a nested sequent calculus on the colored input. Our technique lets us decide the validity of standpoint formulae in CoNP since proof-search only produces a partial proof relative to each permitted coloring of the input. We show how all partial proofs can be fused together to construct a complete proof when the input is valid, and how certain partial proofs can be transformed into a counter-model when the input is invalid. These "certificates" (i.e. proofs and counter-models) serve as explanations of the (in)validity of the input. (shrink)

The debates on the scientificity of social sciences in general, and sociology in particular, are recurring. From the original methodenstreitat the end the 19th Century to the contemporary controversy on the legitimacy of “regional epistemologies”, a same set of interrogations reappears. Are social sciences really scientific? And if so, are they sciences like other sciences? How should we conceive “research programs” Lakatos (1978) or “research traditions” for Laudan (1977) able to produce advancement of knowledge in the field of social and (...) human phenomena? Is the progress of knowledge in social sciences similar to the one generally observed in natural sciences? Is it possible to evaluate the relative merits of each one of these research programs? -/- These debates are important vectors of social and intellectual polarization. The historical divide between the positivist and the hermeneutics poles precedes the structure of the contemporary debate around the epistemic space of social sciences, It is not only a question of renewing the opposition between a monist view of sciences (e.g. McIntyre, 1996) and a dualistic one (e.g. Geertz, 1973) or even a trialist view of sciences (e.g. Lepenies, 1985). It is also a question of asserting dichotomies transformed into framework (including when it is a question of exceeding them): nature-culture, nomothetic-idiographic, models-narrative, structure-history, cause-reason, explanation-comprehension. -/- In this short introduction, we provide, in section 2, a first overview of this epistemological debate in social science. Section 3 proposes a different standpoint on the same questions, by introducing both ontological and methodological aspects in this basic epistemological debate. Namely, following (Hollis, 1994) oppositions of the explanation - understanding, causes - meaning, etc., types discussed in the firsts ection are comparatively examined together with oppositions of the structure - action, holism - individualism types. This allows us to discus show multi-agent design, by integrating various dimensions and standpoints in the same framework (Phan, Amblard, 2007, Chapters 1, 5, 14) can help us to shift these boundaries, and to bypass these oppositions. As model building and ontology design are at the core of this process (Phan, Amblard, 2007, Chapter 12), Section4 discusses various issues of the art of modelling, starting both from economists’ and sociologists’ current standpoints. (shrink)

We consider the problem of executing conscious behavior i.e., of driving an agent’s actions and of allowing it, at the same time, to run concurrent processes reflecting on these actions. Toward this end, we express a single agent’s plans as reflexive dialogs in a multi-agent system defined by a virtual machine. We extend this machine’s planning language by introducing two specific operators for reflexive dialogs i.e., conscious and caught for monitoring beliefs and actions, respectively. The possibility (...) to use the same language both to drive a machine and to establish a reflexive communication within the machine itself stands as a key feature of our model. (shrink)

A dynamic epistemic logic is presented in which the single agent can reason about his knowledge stages before and after announcements. The logic is generated by reinterpreting multi agent private announcements in a single agent environment. It is shown that a knowability principle is valid for such logic: any initially true ϕ can be known after a certain number of announcements.

STIT logic is a prominent framework for the analysis of multi-agent choice-making. In the available deontic extensions of STIT, the principle of Ought-implies-Can (OiC) fulfills a central role. However, in the philosophical literature a variety of alternative OiC interpretations have been proposed and discussed. This paper provides a modular framework for deontic STIT that accounts for a multitude of OiC readings. In particular, we discuss, compare, and formalize ten such readings. We provide sound and complete sequent-style calculi for (...) all of the various STIT logics accommodating these OiC principles. We formally analyze the resulting logics and discuss how the different OiC principles are logically related. In particular, we propose an endorsement principle describing which OiC readings logically commit one to other OiC readings. (shrink)

Readers of Philosophical Psychology may be most familiar with Ron Sun by way of an article recently appearing in this journal on creative composition expressed within his own hybrid computational intelligence model, CLARION (Sun, 2013). That article represents nearly two decades’ work in situated agency stressing the importance of psychologically realistic architectures and processes in the articulation of both functional, and reflectively informative, AI and agent- level social-cultural simulations. Readers may be less familiar with Sun’s 2001 “prolegomena” to related (...) multi-agent (proto-social) research also from this journal. That article argues that “a proper balance between “objective” social reality and individual cognitive processes” is necessary in order to understand “how individual belief systems... and the social/cultural belief system ... interact” (Sun, 2001, pages 10 and 23). This issue remains central in Sun’s 2012 edited volume, Grounding Social Sciences in the Cognitive Sciences, here addressed from within the expanding field of pioneering researchers bent on orchestrating that proper balance, the “cognitive social sciences.” Its fifteen chapters are sectioned according to culture, politics, religion, and economics, and closes with an especially rewarding pair of contributions from Gintis, and McCubbins and Turner, under the heading of “unifying perspectives.” Most entries – but for Sun’s own - are serviceably summarized in the introductory overview. So, rather than follow suit, this review will focus on setting out Sun’s vision, noting how this text helps us to realize it more clearly, with a positive focus on a few entries in particular. (shrink)

An agent exercises instrumental rationality to the degree that she adopts appropriate means to achieving her ends. Adopting appropriate means to achieving one’s ends can, in turn, involve overcoming one’s strongest desires, that is, it can involve exercising synchronic self-control. However, contra prominent approaches, I deny that synchronic self-control is possible. Specifically, I draw on computational models and empirical evidence from cognitive neuroscience to describe a naturalistic, multi-system model of the mind. On this model, synchronic self-control is impossible. (...) Must we, then, give up on a meaningful conception of instrumental rationality? No. A multi-system view still permits something like synchronic self-control: an agent can control her very strong desires. Adopting a multi-system model of the mind thus places limitations on our conceptions of instrumental rationality, without requiring that we abandon the notion altogether. (shrink)

We present cut-free labelled sequent calculi for a central formalism in logics of agency: STIT logics with temporal operators. These include sequent systems for Ldm , Tstit and Xstit. All calculi presented possess essential structural properties such as contraction- and cut-admissibility. The labelled calculi G3Ldm and G3Tstit are shown sound and complete relative to irreflexive temporal frames. Additionally, we extend current results by showing that also Xstit can be characterized through relational frames, omitting the use of BT+AC frames.

The field of value alignment, or more broadly machine ethics, is becoming increasingly important as artificial intelligence developments accelerate. By ‘alignment’ we mean giving a generally intelligent software system the capability to act in ways that are beneficial, or at least minimally harmful, to humans. There are a large number of techniques that are being experimented with, but this work often fails to specify what values exactly we should be aligning. When making a decision, an agent is supposed to (...) maximize the expected utility of its value function. Classically, this has been referred to as happiness, but this is just one of many things that people value. -/- In order to resolve this issue, we need to determine a set of human values that represent humanity's interests. Although this problem might seem intractable, research shows that people of various cultures and religions actually share more in common than they realize. In this book we review world religions, moral philosophy and evolutionary psychology to elucidate a common set of shared values. We then show how these values can be used to address the alignment problem and conclude with problems and goals for future research. -/- The key audience for this book will be researchers in the field of ethics and artificial intelligence who are interested in, or working on this problem. These people will come from various professions and include philosophers, computer programmers and psychologists, as the problem itself is multi-disciplinary. (shrink)

Decision-making under normative uncertainty requires an agent to aggregate the assessments of options given by rival normative theories into a single assessment that tells her what to do in light of her uncertainty. But what if the assessments of rival theories differ not just in their content but in their structure -- e.g., some are merely ordinal while others are cardinal? This paper describes and evaluates three general approaches to this "problem of structural diversity": structural enrichment, structural depletion, and (...) multi-stage aggregation. All three approaches have notable drawbacks, but I tentatively defend multi-stage aggregation as least bad of the three. (shrink)

When sporting agents fail through wrongful or faulty behaviour, they should feel guilty; when they fail because of a deficiency in their abilities, they should feel shame. But sometimes we fail without being deficient and without being at fault. I illustrate this with two examples of players, Moacir Barbosa and Roberto Baggio, who failed in World Cup finals and cost their teams the greatest prize in sport. Although both players failed, I suggest that neither was at fault and neither (...) was deficient. I argue that we can fail through no fault of our own because our abilities are always fallible. This fallibility means that to succeed – to achieve sporting glory – we must run the risk of failure. The appropriate emotion to feel over such failures is agent-regret. Sporting agents and observers should not take up what I call the ‘critical position’: the idea that someone who fails must be deficient or must have been at fault. This allows for a softer, but also more accurate, attitude towards our own failures and the failures of others. I end by suggesting that the fallibility of our abilities is made clear through playing or watching sport, and this can illuminate life more broadly. (shrink)

Agent-regret seems to give rise to a philosophical puzzle. If we grant that we are not morally responsible for consequences outside our control (the ‘Standard View’), then agent-regret—which involves self-reproach and a desire to make amends for consequences outside one’s control—appears rationally indefensible. But despite its apparent indefensibility, agent-regret still seems like a reasonable response to bad moral luck. I argue here that the puzzle can be resolved if we appreciate the role that agent-regret plays in (...) a larger social practice that helps us deal with bad moral luck. That agent-regret is a component in a social practice limits the questions that we can reasonably ask about it. While we can ask whether an experience of agent-regret is rational given the norms of this practice, we cannot ask the question that motivates the puzzle of agent-regret, viz. whether agent-regret is rationally defensible according to the Standard View. (shrink)

Andrew Forcehimes and Luke Semrau argue that agent-relative consequentialism is implausible because in some circumstances it classes an act as impermissible yet holds that the outcome of all agents performing that impermissible act is preferable. I argue that their problem is closely related to Derek Parfit's problem of ‘direct collective self-defeat’ and show how Parfit's plausible solution to his problem can be adapted to solve their problem.

The agent-relative/agent-neutral distinction is one of the most important in contemporary moral theory. Yet, providing an adequate formal account of it has proven difficult. In this article I defend a new formal account of the distinction, one that avoids various problems faced by other accounts. My account is based on an influential account of the distinction developed by McNaughton and Rawling. I argue that their approach is on the right track but that it succumbs to two serious objections. (...) I then show how to formulate a new account that follows the key insights of McNaughton and Rawling’s approach yet avoids the two objections. (shrink)

Intelligence in current AI research is measured according to designer-assigned tasks that lack any relevance for an agent itself. As such, tasks and their evaluation reveal a lot more about our intelligence than the possible intelligence of agents that we design and evaluate. As a possible first step in remedying this, this article introduces the notion of “self-concern,” a property of a complex system that describes its tendency to bring about states that are compatible with its continued self-maintenance. (...) Self-concern, as argued, is the foundation of the kind of basic intelligence found across all biological systems, because it reflects any such system's existential task of continued viability. This article aims to cautiously progress a few steps closer to a better understanding of some necessary organisational conditions that are central to self-concern in biological systems. By emulating these conditions in embodied AI, perhaps something like genuine self-concern can be implemented in machines, bringing AI one step closer to its original goal of emulating human-like intelligence. (shrink)

Toleration classically denotes a relation between two agents that is characterised by three components: objection, power, and acceptance overriding the objection. Against recent claims that classical toleration is not applicable in liberal democracies and that toleration must therefore either be understood purely attitudinally or purely politically, we argue that the components of classical toleration are crucial elements of contemporary cases of minority accommodation. The concept of toleration is applicable to, and is an important element of descriptions of such cases, (...) provided that one views them as wholes, rather than as sets of isolated relations. We explain this by showing how certain cases of toleration are multi-dimensional and how the descriptive concept of toleration might be understood intersectionally. We exemplify this by drawing on case studies of mosque controversies in Germany and Denmark. Finally, we propose that intersectionality is not only relevant to the descriptive concept of toleration but also captures an important aspect of normative theories of toleration. We illustrate this by discussing ideals of respect-based toleration, which we also apply to the case studies. (shrink)