Distributed Artificial Intelligence (DAI) deals with computational systems where several intelligent components interact in a common environment. This paper is aimed at pointing out and fostering the exchange between DAI and cognitive and social science in order to deal with the issues of interaction, and in particular with the reasons and possible strategies for social behaviour in multi-agent interaction is also described which is motivated by requirements of cognitive plausibility and grounded the notions of power, dependence and help. (...) Connections with human-computer interaction are also suggested. (shrink)

In current philosophical research the term 'philosophy of social action' can be used - and has been used - in a broad sense to encompass the following central research topics: 1) action occurring in a social context; this includes multi-agent action; 2) joint attitudes (or "we-attitudes" such as joint intention, mutual belief) and other social attitudes needed for the explication and explanation of social action; 3) social macro-notions, such as actions performed by social groups and properties of social groups such (...) as their goals and beliefs; 4) social norms and social institutions (see Tuomela, 1984, 1995). The theory of social action understood analogously in a broad sense would then involve not only philosophical but all other relevant theorizing about social action. Thus, in this sense, such fields of Artificial Intelligence (AI) as Distributed AI (DAI) and the theory of Multi-Agent Systems (MAS) fall within the scope of the theory of social action. DAI studies the social side of computer systems and includes various well-known areas ranging from Human Computer Interaction, Computer-Supported Cooperative Work, Organizational Processing, Distributed Problem Solving to Simulation of Social Systems and Organizations. Even if I am a philosopher with low artificial intelligence I will below try to say something about what the scope of DAI should be taken to be on conceptual and philosophical grounds. (In the later sections of the paper the central notion of joint intention will be the main topic - in order to illustrate how philosophers and DAI-researchers approach this issue.) Let us now consider the relationship between philosophy - especially philosophy of social action - and DAI. Both are concerned with social matters and in this sense seem to have a connection to social science proper. What kinds of questions should these areas of study be concerned with? In principle, ordinary social science should study all aspects of social life (in various societies and cultures), try to describe it and create general theories to explain it. (shrink)

This paper studies the traditional target classification and recognition algorithm based on Histogram of Oriented Gradients feature extraction and Support Vector Machine classification and applies this algorithm to distributed artificial intelligence image recognition. Due to the huge number of images, the general detection speed cannot meet the requirements. We have improved the HOG feature extraction algorithm. Using principal component analysis to perform dimensionality reduction operations on HOG features and doing distributed artificial intelligence image recognition experiments, the (...) results show that the image detection efficiency is slightly improved, and the detection speed is also improved. This article analyzes the reason for these changes because PCA mainly uses the useful feature information in HOG features. The parallelization processing of HOG features on graphics processing unit is studied. GPU is used for high parallel and high-density calculations, and the calculation of HOG features is very complicated. Using GPU for parallelization of HOG features can make the calculation speed of HOG features improved. We use image experiments for the parallelized HOG feature algorithm. Experimental simulations show that the speed of distributed artificial intelligence image recognition is greatly improved. By analyzing the existing digital image recognition methods, an improved BP neural network algorithm is proposed. Under the premise of ensuring accuracy, the recognition speed of digital images is accelerated, the time required for recognition is reduced, real-time performance is guaranteed, and the effectiveness of the algorithm is verified. (shrink)

A series of recent debates in experimental economics have associated demand effects with the artificiality of the experimental setting and have linked it to the problem of external validity. In this paper, we argue that these associations can be misleading, partly because of the ambiguity with which “artificiality” has been defined, but also because demand effects and external validity are related in complex ways. We argue that artificiality may be directly as well as inversely correlated with demand effects. We also (...) distinguish between the demand effects of experimentation and the reactions that they may trigger and that might endanger experimental validity. We conclude that economists should pay more attention to the way in which subjects construe the experimental task and learn to exploit subjects’ reactivity to expectations in their experiments. (shrink)

In this paper, a proposed modified stochastic fractal search algorithm is applied to find the most appropriate site and size of capacitor banks for distribution systems with 33, 69, and 85 buses. Two single-objective functions are considered to be reduction of power loss and reduction of total cost of energy loss and capacitor investment while satisfying limit of capacitors, limit of conductor, and power balance of the systems. MSFS was developed by performing three new mechanisms including new diffusion mechanism and (...) two new update mechanisms on the conventional stochastic fractal search algorithm. As a result, MSFS can reduce 0.002%, 0.003%, and 0.18% of the total power loss from SFS for the three study systems. As compared to other methods, MSFS can reduce power loss from 0.07% to 3.98% for the first system, from 3.7% to 7.3% for the second system, and from 0.92% to 6.98% for the third system. For the reduction of total cost, the improvement level of the proposed method over SFS and two other methods is more significant. It is 0.03%, 1.22%, and 5.76% for the second system and 2.31%, 0.87%, and 3.77% for the third system. It is emphasized that the proposed method can find the global optimal solutions for all study cases while SFS was still implementing search process nearby or far away from the solutions. Furthermore, MSFS can converge to the best solutions much faster than these compared methods. Consequently, it can be concluded that the proposed method is very effective for finding the best location and size of added capacitors in distribution power systems. (shrink)

Page argues that localist models can be applied to a number of problems that are difficult for distributed models. However, it is easy to find examples where the opposite is true. This commentary illustrates the superiority of distributed models in the domain of artificial grammar learning, a paradigm widely used to investigate implicit learning.

While feminist critiques of AI are increasingly common in the scholarly literature, they are by no means new. Alison Adam’s Artificial Knowing (1998) brought a feminist social and epistemological stance to the analysis of AI, critiquing the symbolic AI systems of her day and proposing constructive alternatives. In this paper, we seek to revisit and renew Adam’s arguments and methodology, exploring their resonances with current feminist concerns and their relevance to contemporary machine learning. Like Adam, we ask how new (...) AI methods could be adapted for feminist purposes and what role new technologies might play in addressing concerns raised by feminist epistemologists and theorists about algorithmic systems. In particular, we highlight distributed and federated learning as providing partial solutions to the power-oriented concerns that have stymied efforts to make machine learning systems more representative and pluralist. (shrink)

Artificial Intelligence (AI) systems are ubiquitous. From social media timelines, video recommendations on YouTube, and the kinds of adverts we see online, AI, in a very real sense, filters the world we see. More than that, AI is being embedded in agent-like systems, which might prompt certain reactions from users. Specifically, we might find ourselves feeling frustrated if these systems do not meet our expectations. In normal situations, this might be fine, but with the ever increasing sophistication of AI-systems, (...) this might become a problem. While it seems unproblematic to realize that being angry at your car for breaking down is unfitting, can the same be said for AI-systems? In this paper, therefore, I will investigate the so-called “reactive attitudes”, and their important link to our responsibility practices. I then show how within this framework there exist exemption and excuse conditions, and test whether our adopting the “objective attitude” toward agential AI is justified. I argue that such an attitude is appropriate in the context of three distinct senses of responsibility (answerability, attributability, and accountability), and that, therefore, AI-systems do not undermine our responsibility ascriptions. (shrink)

In lieu of an abstract, here is a brief excerpt of the content:Commentary on “Towards a Design-Based Analysis of Emotional Episodes”Cristiano Castelfranchi (bio) and Maria Miceli (bio)Keywordsgrief, suffering, attachment, agent architectureThis paper is significant in many respects: its approach (the design-based analysis); its proposed architecture; its description of grief; and its self-control/perturbance theory. We would offer some remarks on each of these aspects.AI: Back to the FutureAfter some years of crisis, AI seems now to have recovered its original challenging attitude (...) and its identity. In particular all the original ambitions of AI have been renewed under the label of “autonomous agents” (see, for example, Russell and Norvig 1995; Wooldridge and Jennings 1995). This paper and its authors are good representatives of this challenging attitude. Of course, the AI appearing at century’s end is different from, say, the AI of the 1980s. AI systems and models are now conceived for reasoning and acting in open unpredictable worlds, with limited and uncertain knowledge, in real time, with bounded (both cognitive and material) resources, with hybrid architectures, interfering—either co-operatively or competitively—with other systems. The new password for the AI of the 1990s is interaction (Bobrow 1991): interaction with an evolving environment; among several, distributed and heterogeneous artificial systems in a network; with human users; and among humans through computers.This renewed AI, however different though it is in some ways, can be understood as returning to its original aim of modeling “intelligence”: not only static and isolated sub-functions of intelligence but the integrated and pragmatic functioning of autonomous agent architectures as against algorithms. Our renewed AI, based on architectures, can again have important exchanges with other (psychological and social) sciences. This paper is a good example of the fruitful cross-fertilization of ideas which may result from such exchanges. It was indeed just this which originally gave rise to cognitive science.The ArchitectureThe proposed architecture, although quite speculative, and—as the authors themselves emphasize—very much a first sketch, is directly relevant to this new AI. In fact, recent research efforts in the agent-architecture domain are focused exactly on the problem of “goal dynamics” [End Page 129] or “motive processing,” which is the core of this architecture, and on the problem of integrating reactivity and deliberation. Compared with current AI architectures (for example, BDI systems) it is important to note that this “motive-processing” architecture is not yet formalized or implemented to the same extent. On the other hand, it is more interesting from a psychological point of view, and is already able to give us some important insights into human emotions, in particular those emotions (like grief) that involve all or total loss of control of management processes.The Report-Based Analysis of GriefThe report-based description of grief provides a persuasive and rich basis for both modeling and discussing the model. The relevant features of this emotion are well characterized with reference to the self-control problem and motive-processing (though, as we shall see, to use grief as an example of those emotions that involve all or total loss of control of thought processes may provide a partially distorted view of grief itself). In particular, we would stress the relevance of the “intrusive” character of grief-related contents; the rejection of certain beliefs due to the psychic pain they cause; the coping strategies people describe; and the second order motivations related to the control of the emotional state. We believe that psychological models especially of these aspects are usually less specific and explanatory.Let us now try to synthesize some criticisms one could advance of this analysis. We will first address two core notions of the model—perturbance and self-control—and then consider more specifically the treatment of grief.PerturbanceThis notion is potentially very interesting, unless it is reduced to an information processing perspective (e.g., perturbance of attentive processes). An information processing perspective is much more limited than a motive-processing one—at least in the limitative sense of the term, as it is generally used in the psychology of cognitive processes—and is in our view insufficient for understanding human emotions and their role. This limiting interpretation of perturbance fails... (shrink)

This paper provides a new analysis of e - trust , trust occurring in digital contexts, among the artificial agents of a distributed artificial system. The analysis endorses a non-psychological approach and rests on a Kantian regulative ideal of a rational agent, able to choose the best option for itself, given a specific scenario and a goal to achieve. The paper first introduces e-trust describing its relevance for the contemporary society and then presents a new theoretical analysis (...) of this phenomenon. The analysis first focuses on an agent’s trustworthiness , this one is presented as the necessary requirement for e-trust to occur. Then, a new definition of e-trust as a second-order-property of first-order relations is presented. It is shown that the second-order-property of e-trust has the effect of minimising an agent’s effort and commitment in the achievement of a given goal. On this basis, a method is provided for the objective assessment of the levels of e-trust occurring among the artificial agents of a distributed artificial system. (shrink)

Artificial agents have become increasingly prevalent in human social life. In light of the diversity of new human–machine interactions, we face renewed questions about the distribution of moral responsibility. Besides positions denying the mere possibility of attributing moral responsibility to artificial systems, recent approaches discuss the circumstances under which artificial agents may qualify as moral agents. This paper revisits the discussion of how responsibility might be distributed between artificial agents and human interaction partners (including producers (...) of artificial agents) and raises the question of whether attributions of responsibility should remain entirely on the human side. While acknowledging a crucial difference between living human beings and artificial systems culminating in an asymmetric feature of human–machine interactions, this paper investigates the extent to which artificial agents may reasonably be attributed a share of moral responsibility. To elaborate on criteria that can justify a distribution of responsibility in certain human–machine interactions, the role of types of criteria (interaction-related criteria and criteria that can be deferred from socially constructed responsibility relationships) is examined. Thereby, the focus will lay on the evaluation of potential criteria referring to the fact that artificial agents surpass in some aspects the capacities of humans. This is contrasted with socially constructed responsibility relationships that do not take these criteria into account. In summary, situations are examined in which it seems plausible that moral responsibility can be distributed between artificial and human agents. (shrink)

Various arguments have been provided for drawing non-humans such as animals and artificial agents into the sphere of moral consideration. In this paper, I argue for a shift from an ontological to a social-philosophical approach: instead of asking what an entity is, we should try to conceptually grasp the quasi-social dimension of relations between non-humans and humans. This allows me to reconsider the problem of justice, in particular distributive justice . Engaging with the work of Rawls, I show that (...) an expansion of the contractarian framework to non-humans causes an important problem for liberalism, but can be justified by a contractarian argument. Responding to Bell’s and Nussbaum’s comments on Rawls, I argue that we can justify drawing non-humans into the sphere of distributive justice by relying on the notion of a co-operative scheme. I discuss what co-operation between humans and non-humans can mean and the extent to which it depends on properties. I conclude that we need to imagine principles of ecological and technological distributive justice. (shrink)

Extreme events, which are usually characterized by generalized extreme value models, can exhibit long-term memory, whose impact needs to be quantified. It was known that extreme recurrence intervals can better characterize the significant influence of long-term memory than using the GEV model. Our statistical analyses based on time series datasets following the Lévy stable distribution confirm that the stretched exponential distribution can describe a wide spectrum of memory behavior transition from exponentially distributed intervals to power-law distributed ones, extending (...) the previous evaluation of the stretched exponential function using Gaussian/exponential distributed random data. Further deviation and discussion of a historical paradox are also provided, based on the theoretical analysis of the Bayesian law and the stretched exponential distribution. (shrink)

Presupposing no familiarity with the technical concepts of either philosophy or computing, this clear introduction reviews the progress made in AI since the inception of the field in 1956. Copeland goes on to analyze what those working in AI must achieve before they can claim to have built a thinking machine and appraises their prospects of succeeding. There are clear introductions to connectionism and to the language of thought hypothesis which weave together material from philosophy, artificial intelligence and neuroscience. (...) John Searle's attacks on AI and cognitive science are countered and close attention is given to foundational issues, including the nature of computation, Turing Machines, the Church-Turing Thesis and the difference between classical symbol processing and parallel distributed processing. The book also explores the possibility of machines having free will and consciousness and concludes with a discussion of in what sense the human brain may be a computer. (shrink)

Is equality a distributive value or does it rather point to the quality of social relationships? This article criticizes the distributive character of luck egalitarian theories of justice and fleshes out the central characteristics of an alternative, relational approach to equality. It examines a central objection to distributive theories: that such theories cannot account for the significance of how institutions treat people (as opposed to the outcomes they bring about). I discuss two variants of this objection: first, that distributive theories (...) fail to account for the importance of how institutions cause good shortfalls and, second, that they fail to account for the normative attitude of social and political institutions expressed in different ways of treating people. The article argues that the causal variant of the objection has only very limited reach, and endorses the expressive variant: the attitudes expressed by institutions in their treatment of persons, such as contempt or neglect, generate potentially unjust social relationships and hierarchies. This should be the focus of a relational egalitarian approach to social justice. The article proceeds to explain how it is possible that artificial agents such as institutions have attitudes and how these attitudes are not reducible to those of the individuals that sustain them, and argues that distributive theories cannot be so modified as to account for such attitudes. It concludes by indicating several directions for the development of more worked-out conceptions of relational equality on this basis. (shrink)

While the term “reactivity” has come to be associated with specific phenomena in the social sciences, having to do with subjects’ awareness of being studied, this paper takes a broader stance on this concept. I argue that reactivity is a ubiquitous feature of the psychological subject matter and that this fact is a precondition of experimental research, while also posing potential problems for the experimenter. The latter are connected to the worry about distorted data and experimental artifacts. But what are (...) experimental artifacts and what is the most productive way of dealing with them? In this paper, I approach these questions by exploring the ways in which experimenters in psychology simultaneously exploit and suppress the reactivity of their subject matter in order to produce experimental data that speak to the question or subject matter at hand. Highlighting the artificiality of experimental data, I raise the question of what distinguishes a genuine experimental result from an experimental artifact. My analysis construes experimental results as the outcomes of inferences from the data that take material background assumptions as auxiliary premises. Artifacts occur when one or more of these background assumptions are false, such that the data do not reliably serve the purposes they were generated for. I conclude by laying out the ways in which my analysis of data quality is relevant to, and informed by, recent debates about the replicability of experimental results. (shrink)

Presupposing no familiarity with the technical concepts of either philosophy or computing, this clear introduction reviews the progress made in AI since the inception of the field in 1956. Copeland goes on to analyze what those working in AI must achieve before they can claim to have built a thinking machine and appraises their prospects of succeeding.There are clear introductions to connectionism and to the language of thought hypothesis which weave together material from philosophy, artificial intelligence and neuroscience. John (...) Searle's attacks on AI and cognitive science are countered and close attention is given to foundational issues, including the nature of computation, Turing Machines, the Church-Turing Thesis and the difference between classical symbol processing and parallel distributed processing. The book also explores the possibility of machines having free will and consciousness and concludes with a discussion of in what sense the human brain may be a computer. (shrink)

Open peer commentary on the article “Exploration of the Functional Properties of Interaction: Computer Models and Pointers for Theory” by Etienne B. Roesch, Matthew Spencer, Slawomir J. Nasuto, Thomas Tanay & J. Mark Bishop. Upshot: Although the authors investigate a form of distributed swarm intelligence and solve some problems with it – including sorting and summing – the major goal, which is constructing cognition, cannot be achieved by this approach alone. I propose that anticipatory mechanisms have the potential to (...) construct cognition and may very well be combined with swarm intelligence principles. (shrink)

Accountability is a cornerstone of the governance of artificial intelligence (AI). However, it is often defined too imprecisely because its multifaceted nature and the sociotechnical structure of AI systems imply a variety of values, practices, and measures to which accountability in AI can refer. We address this lack of clarity by defining accountability in terms of answerability, identifying three conditions of possibility (authority recognition, interrogation, and limitation of power), and an architecture of seven features (context, range, agent, forum, standards, (...) process, and implications). We analyse this architecture through four accountability goals (compliance, report, oversight, and enforcement). We argue that these goals are often complementary and that policy-makers emphasise or prioritise some over others depending on the proactive or reactive use of accountability and the missions of AI governance. (shrink)

The computer simulation study explores the impact of the duration of social impact on the generation and stabilization of cooperative strategies. Rather than seeding the simulations with a finite set of strategies, a continuous distribution of strategies is being defined. Members of heterogeneous populations were characterized by a pair of probabilistic reactive strategies: the probability to respond to cooperation by cooperation and the probability to respond to defection by cooperation. This generalized reactive strategy yields the standard TFT mechanism, (...) the All-Cooperate, All-Defect and Bully strategies as special cases. Pairs of strategies interacted through a Prisoner's Dilemma game and exerted social influence on all other members. Manipulating: (i) the initial distribution of populations' strategies, and (ii) the duration of social influence, we monitored the conditions leading to the emergence and stabilization of cooperative strategies. Results show that: (1) The duration of interactions between pairs of strategies constitutes a crucial factor for the emergence and stabilization of cooperative strategies, (2) Unless sufficient learning intervals are provided, initializing the simulations with cooperative populations does not guarantee that cooperation will sustain. (shrink)

Distributed cognition is widely recognized as an approach to the study of all cognition. It identifies the distribution of cognitive processes between persons and technology, among people, and across time in the development of the social and material contexts for thinking. This paper suggests an ectoderm-centric perspective as the basis for distributed cognition, and in so doing redefines distributed cognition as the ability of an organism to interact with its environment for the purpose of satisfying its most (...) basic physiological (internal and external) and social needs in order to survive and sustain itself. Underlying this ectoderm-centric perspective is a proposed theory of reactive and interactive learning. (shrink)

Buildings energy consumption is growing gradually and put away around 40% of total energy use. Predicting heating and cooling loads of a building in the initial phase of the design to find out optimal solutions amongst different designs is very important, as ell as in the operating phase after the building has been finished for efficient energy. In this study, an artificial neural network model was designed and developed for predicting heating and cooling loads of a building based on (...) a dataset for building energy performance. The main factors for input variables are: relative compactness, roof area, overall height, surface area, glazing are a, wall area, glazing area distribution of a building, orientation, and the output variables: heating and cooling loads of the building. The dataset used for training are the data published in the literature for various 768 residential buildings. The model was trained and validated, most important factors affecting heating load and cooling load are identified, and the accuracy for the validation was 99.60%. (shrink)

Presents a series of generalizations of distributivity operators across a type hierarchy, in order to account for collective-distributive ambiguities for non-subject arguments.

There is, in some quarters, concern about high–level machine intelligence and superintelligent AI coming up in a few decades, bringing with it significant risks for humanity. In other quarters, these issues are ignored or considered science fiction. We wanted to clarify what the distribution of opinions actually is, what probability the best experts currently assign to high–level machine intelligence coming up within a particular time–frame, which risks they see with that development, and how fast they see these developing. We thus (...) designed a brief questionnaire and distributed it to four groups of experts in 2012/2013. The median estimate of respondents was for a one in two chance that high-level machine intelligence will be developed around 2040-2050, rising to a nine in ten chance by 2075. Experts expect that systems will move on to superintelligence in less than 30 years thereafter. They estimate the chance is about one in three that this development turns out to be ‘bad’ or ‘extremely bad’ for humanity. (shrink)