Social media platforms are known to optimize user engagement with the help of algorithms. It is widely understood that this practice gives rise to echo chambers - users are mainly exposed to opinions that are similar to their own. In this paper, we ask whether echo chambers are an inevitable result of high engagement; we address this question in a novel model. Our main theoretical results establish bounds on the maximum engagement achievable under a diversity constraint, for suitable measures of engagement and diversity; we can therefore quantify the worst-case tradeoff between these two objectives. Our empirical results, based on real data from Twitter, chart the Pareto frontier of the engagement-diversity tradeoff.

Abstract Constraining the actions of AI systems is one promising way to ensure that these systems behave in a way that is morally acceptable to humans. But constraints alone come with drawbacks as in many AI systems, they are not flexible. If these constraints are too rigid, they can preclude actions that are actually acceptable in certain, contextual situations. Humans, on the other hand, can often decide when a simple and seemingly inflexible rule should actually be overridden based on the context. In this paper, we empirically investigate the way humans make these contextual moral judgements, with the goal of building AI systems that understand when to follow and when to override constraints. We propose a novel and general preference-based graphical model that captures a modification of standard dual process theories of moral judgment. We then detail the design, implementation, and results of a study of human participants who judge whether it is acceptable to break a well-established rule: no cutting in line. We then develop an instance of our model and compare its performance to that of standard machine learning approaches on the task of predicting the behavior of human participants in the study, showing that our preference-based approach more accurately captures the judgments of human decision-makers. It also provides a flexible method to model the relationship between variables for moral decision-making tasks that can be generalized to other settings.

Fear induction in the form of stories and visual images pervades the history of human culture. Creating a visceral emotion such as fear remains one of the cornerstones of human creativity. As artificial intelligence makes strides in solving challenging analytical problems like chess and Go, an important question still remains: can machines induce extreme human emotions, such as fear? In this work, we propose a deep-learning based collaborative horror writer that collaboratively writes scary stories with people on Twitter. We deploy our system as a bot on Twitter that regularly generates and posts new stories on Twitter, and invites users to participate. Users who interact with the stories produce multiple storylines originating from the same tweet, thereby creating a tree-based story structure. We further perform a validation study on n = 105 subjects to verify whether the generated stories psychologically move people on psychometrically validated measures of effect and anxiety such as I-PANAS-SF [43] and STAI-SF [26]. Our experiments show that 1) stories generated by our bot as well as the stories generated collaboratively between our bot and Twitter users produced statistically significant increases in negative affect and state anxiety compared to the control condition, and 2) collaborated stories are more successful in terms of increasing negative affect and state anxiety than the machine-generated ones. Furthermore, we make three novel datasets used in our framework publicly available at https://github.com/catlab-team/shelley for encouraging further research on this topic.

Superintelligence is a hypothetical agent that possesses intelligence far surpassing that of the brightest and most gifted human minds. In light of recent advances in machine intelligence, a number of scientists, philosophers and technologists have revived the discussion about the potentially catastrophic risks entailed by such an entity. In this article, we trace the origins and development of the neo-fear of superintelligence, and some of the major proposals for its containment. We argue that total containment is, in principle, impossible, due to fundamental limits inherent to computing itself. Assuming that a superintelligence will contain a program that includes all the programs that can be executed by a universal Turing machine on input potentially as complex as the state of the world, strict containment requires simulations of such a program, something theoretically (and practically) impossible. This article is part of the special track on AI and Society.

Bitcoin is a cryptocurrency and a decentralized semi-anonymous peer-to-peer payment system in which the transactions are verified by network nodes and recorded in a public massively replicated ledger called the blockchain. Bitcoin is currently considered as one of the most disruptive technologies. Bitcoin represents a paradox of opposing forces. On one hand, it is fundamentally social, allowing people to transact in a peer-to-peer manner to create and exchange value. On the other hand, Bitcoin's core design philosophy and user base contain strong anti-social elements and constraints, emphasizing anonymity, privacy, and subversion of traditional centralized financial systems. We believe that the success of Bitcoin, and the financial ecosystem built around it, will likely rely on achieving an optimal balance between these social and anti-social forces. To elucidate the role of these forces, we analyze the evolution of the entire Bitcoin transaction graph from its inception, and quantify the evolution of its key structural properties. We observe that despite its different nature, the Bitcoin transaction graph exhibits many universal dynamics typical of social networks. However, we also find that Bitcoin deviates in important ways due to anonymity-seeking behavioral patterns of its users. As a result, the network exhibits a two-orders-of-magnitude larger diameter, sparse treelike communities, and an overwhelming majority of transitional or intermediate accounts with incoming and outgoing edges but zero cumulative balances. These results illuminate the evolutionary dynamics of the most popular cryptocurrency, and provide us with initial understanding of social networks rooted in and driven by anti-social constraints.

Everyone agrees that autonomous cars ought to save lives. Even if the cars do not live up to the most optimistic estimates of eliminating 90% of traffic fatalities [1], eliminating at least some traffic fatalities is one of the key promises of automated driving. Indeed, the first two principles of the German Ethics Code for Automated and Connected Vehicles lead with this goal as a normative imperative [2].The primary purpose of partly and fully automated transport systems is to improve safety for all road users. The licensing of automated systems is not justifiable unless it promises to produce at least a diminution in harm compared with human driving [...].

We present a general approach to automating ethical decisions, drawing on machine learning and computational social choice. In a nutshell, we propose to learn a model of societal preferences, and, when faced with a specific ethical dilemma at runtime, efficiently aggregate those preferences to identify a desirable choice. We provide a concrete algorithm that instantiates our approach; some of its crucial steps are informed by a new theory of swap-dominance efficient voting rules. Finally, we implement and evaluate a system for ethical decision making in the autonomous vehicle domain, using preference data collected from 1.3 million people through the Moral Machine website.

We introduce a computational model for building moral autonomous vehicles by learning and generalizing from human moral judgments. We draw on a cognitively inspired model of how people and young children learn moral theories from sparse and noisy data and integrate observations made from different people in different groups. The problem of moral learning for autonomous vehicles is cast as learning how to weigh the different features of the dilemma using utility calculus, with the goal of making these trade-offs reflect how people make them in a wide variety of moral dilemma. By modeling the structures of individuals and groups in a hierarchical Bayesian model, we show that an individual's moral values -- as well as a group's shared values -- can be inferred from sparse and noisy data. We evaluate our approach with data from the Moral Machine, a web application that collects human judgments on moral dilemmas involving autonomous vehicles, and show that the model rapidly and accurately infers people's preferences and can predict the difficulty of moral dilemmas from limited data.

This paper briefly describes AIF-EL, an OWL2-EL compliant ontology for the Argument Interchange Format.

We introduce TuringBox, a platform to democratize the study of AI. On one side of the platform, AI contributors upload existing and novel algorithms to be studied scientifically by others. On the other side, AI examiners develop and post machine intelligence tasks to evaluate and characterize the outputs of algorithms. We outline the architecture of such a platform, and describe two interactive case studies of algorithmic auditing on the platform.

The analysis of the creation, mutation, and propagation of social media content on the Internet is an essential problem in computational social science, affecting areas ranging from marketing to political mobilization. A first step towards understanding the evolution of images online is the analysis of rapidly modifying and propagating memetic imagery or "memes". However, a pitfall in proceeding with such an investigation is the current incapability to produce a robust semantic space for such imagery, capable of understanding differences in Image Macros. In this study, we provide a first step in the systematic study of image evolution on the Internet, by proposing an algorithm based on sparse representations and deep learning to decouple various types of content in such images and produce a rich semantic embedding. We demonstrate the benefits of our approach on a variety of tasks pertaining to memes and Image Macros, such as image clustering, image retrieval, topic prediction and virality prediction, surpassing the existing methods on each. In addition to its utility on quantitative tasks, our method opens up the possibility of obtaining the first large-scale understanding of the evolution and propagation of memetic imagery.

NLP tasks are often limited by scarcity of manually annotated data. In social media sentiment analysis and related tasks, researchers have therefore used binarized emoticons and specific hashtags as forms of distant supervision. Our paper shows that by extending the distant supervision to a more diverse set of noisy labels, the models can learn richer representations. Through emoji prediction on a dataset of 1246 million tweets containing one of 64 common emojis we obtain state-of-the-art performance on 8 benchmark datasets within sentiment, emotion and sarcasm detection using a single pretrained model. Our analyses confirm that the diversity of our emotional labels yield a performance improvement over previous distant supervision approaches.

Highly automated robot ecologies (HARE), or societies of independent autonomous robots or agents, are rapidly becoming an important part of much of the world's critical infrastructure. As with human societies, regulation, wherein a governing body designs rules and processes for the society, plays an important role in ensuring that HARE meet societal objectives. However, to date, a careful study of interactions between a regulator and HARE is lacking. In this paper, we report on three user studies which give insights into how to design systems that allow people, acting as the regulatory authority, to effectively interact with HARE. As in the study of political systems in which governments regulate human societies, our studies analyze how interactions between HARE and regulators are impacted by regulatory power and individual (robot or agent) autonomy. Our results show that regulator power, decision support, and adaptive autonomy can each diminish the social welfare of HARE, and hint at how these seemingly desirable mechanisms can be designed so that they become part of successful HARE.

Given a set of conflicting arguments, there can exist multiple plausible opinions about which arguments should be accepted, rejected or deemed undecided. We study the problem of how multiple such judgements can be aggregated. We define the problem by adapting various classical social-choice-theoretic properties for the argumentation domain. We show that while argument-wise plurality voting satisfies many properties, it fails to guarantee the collective rationality of the outcome. We then present more general results, proving multiple impossibility results on the existence of any good aggregation operator. After characterizing the sufficient and necessary conditions for satisfying collective rationality, we study whether restricting the domain of argument-wise plurality voting to classical semantics allows us to escape the impossibility result. We close by mentioning a couple of graph-theoretical restrictions under which the argument-wise plurality rule does produce collectively rational outcomes. In addition to identifying fundamental barriers to collective argument evaluation, our results contribute to research at the intersection of the argumentation and computational social choice fields.

An inconsistent knowledge base can be abstracted as a set of arguments and a defeat relation among them. There can be more than one consistent way to evaluate such an argumentation graph. Collective argument evaluation is the problem of aggregating the opinions of multiple agents on how a given set of arguments should be evaluated. It is crucial not only to ensure that the outcome is logically consistent, but also satisfies measures of social optimality and immunity to strategic manipulation. This is because agents have their individual preferences about what the outcome ought to be. In the current paper, we analyze three previously introduced argument-based aggregation operators with respect to Pareto optimality and strategy-proofness under different general classes of agent preferences. We highlight fundamental trade-offs between strategic manipulability and social optimality on one hand, and classical logical criteria on the other. Our results motivate further investigation into the relationship between social choice and argumentation theory. The results are also relevant for choosing an appropriate aggregation operator given the criteria that are considered more important, as well as the nature of agents’ preferences.

On the Web, there is always a need to aggregate opinions from the crowd (as in posts, social networks, forums, etc.). Different mechanisms have been implemented to capture these opinions such as Like in Facebook, Favorite in Twitter, thumbs-up/-down, flagging, and so on. However, in more contested domains (e.g., Wikipedia, political discussion, and climate change discussion), these mechanisms are not sufficient, since they only deal with each issue independently without considering the relationships between different claims. We can view a set of conflicting arguments as a graph in which the nodes represent arguments and the arcs between these nodes represent the defeat relation. A group of people can then collectively evaluate such graphs. To do this, the group must use a rule to aggregate their individual opinions about the entire argument graph. Here we present the first experimental evaluation of different principles commonly employed by aggregation rules presented in the literature. We use randomized controlled experiments to investigate which principles people consider better at aggregating opinions under different conditions. Our analysis reveals a number of factors, not captured by traditional formal models, that play an important role in determining the efficacy of aggregation. These results help bring formal models of argumentation closer to real-world application.

In an abstract argumentation framework, there are often multiple plausible ways to evaluate (or label) the status of each argument as accepted, rejected, or undecided. But often there exists a critical set of arguments whose status is sufficient to determine uniquely the status of every other argument. Once an agent has decided its position on a critical set of arguments, then essentially the entire frame-work has been evaluated. Likewise, once a group, e.g. a jury, agrees on the status of a critical set of arguments, all of their different views over all other arguments are resolved. Thus, critical sets of arguments are important both for efficient evaluation by individual agents and for collective agreement by groups of such. To exploit this idea in practice, however, a number of computational questions must be considered. In particular, how much computational effort is needed to verify that a set is, indeed, a critical set or a minimal critical set. In this paper we determine exact bounds on the computational complexity of these and related questions. In addition we provide similar analyses of issues: a concept closely related to critical set and derived in terms of (equivalence) classes of arguments related through “common” labelling behaviours.

Networks are seen everywhere in our modern life, including the Internet, the Grid, P2P file sharing, and sensor networks. Consequently, researchers in Artificial Intelligence (and Multi-Agent Systems in particular) have been actively seeking methods for optimizing the performance of these networks. A promising yet challenging optimization direction is multi-agent learning: allowing agents to adapt their behavior through interaction with one another. However, understanding the dynamics of an adaptive agent network is complicated due to the large number of system parameters, the concurrency by which the system parameters change, and the delay in the effect/consequence of parameter changes. All these factors make it hard to understand why an adaptive network of agents performed well at some time and poorly at another. In this paper we present a software tool that enables researchers in the multi-agent systems field to visualize and analyze the evolution of adaptive networks. The proposed software customizes and implements techniques from data mining and social network analysis research and augment these techniques in order to analyze local agent behaviors. We use our tool to analyze two domains. In both domains we are able to report and explain interesting observations using our tool.

Recently Argumentation Mechanism Design (ArgMD) was introduced as a paradigm for studying argumentation using game-theoretic techniques. To date, this framework has been used to study under what conditions a direct mechanism based on Dung's grounded semantics is strategy-proof (i.e. truth-enforcing) when knowledge of arguments is private to self-interested agents. In this paper, we study Dung's preferred semantics in order to understand under what conditions it is possible to design strategy-proof mechanisms. This is challenging since, unlike with the grounded semantics, there may be multiple preferred extensions, forcing a mechanism to select one. We show that this gives rise to interesting strategic behaviour, and we show that in general it is not possible to have a strategy-proof mechanism that selects amongst the preferred extensions in a non-biased manner. We also investigaet refinements of preferred semantics which induce unique outcomes, namely the skeptical-preferred and ideal semantics.

A major challenge in the field of Multi-Agent Systems is to enable autonomous agents to allocate tasks efficiently. This paper extends previous work on an approach to the collective iterative allocation problem where a group of agents endeavours to f

While argumentation-based negotiation has been accepted as a promising alternative to game-theoretic or heuristic based negotiation, no evidence has been provided to confirm this theoretical advantage. We propose a model of bilateral negotiation extending a simple monotonic concession protocol by allowing the agents to exchange information about their underlying interests and possible alternatives to achieve them during the negotiation. We present an empirical study that demonstrates (through simulation) the advantages of this interest-based negotiation approach over the more classic monotonic concession approach to negotiation.