The topic of value alignment in AI has been gaining significant attention. The ultimate objective is how AI systems can align with human values. One of the main challenges, however, is identifying the relevant values. Some emerging works are focusing on learning relevant values through analysing our statements on social media. In most of these cases, learned values are simply specified through a label such as equality, fairness, or justice. However, the semantics of these values is not studied further. Addressing this gap, we propose a user study that provides us with a systematic approach for learning value semantics. The study is in the context of healthcare and the values of interest are the four fundamental bioethical principles: beneficence, non-maleficence, autonomy, and justice. We conduct a user study to collect the views of medical professionals on the value alignment of synthetic patient cases. We then search the space of candidate functions to find the one that best fits the participants answers, and hence, best describes their view of the value in question.

In the last decades, there has been a deceleration in the rates of poverty reduction, suggesting that traditional redistributive approaches to poverty mitigation could be losing effectiveness, and alternative insights to advance the number one UN Sustainable Development Goal are required. The criminalization of poor people has been denounced by several NGOs, and an increasing number of voices suggest that discrimination against the poor (a phenomenon known as aporophobia) could be an impediment to mitigating poverty. In this paper, we present the novel Aporophobia Agent-Based Model (AABM) to provide evidence of the correlation between aporophobia and poverty computationally. We present our use case built with real-world demographic data and poverty-mitigation public policies (either enforced or under parliamentary discussion) for the city of Barcelona. We classify policies as discriminatory or non-discriminatory against the poor, with the support of specialized NGOs, and we observe the results in the AABM in terms of the impact on wealth inequality. The simulation provides evidence of the relationship between aporophobia and the increase of wealth inequality levels, paving the way for a new generation of poverty reduction policies that act on discrimination and tackle poverty as a societal problem (not only a problem of the poor).

One of the possible ways to embed values into autonomous agents is through reasoning over the norms that govern the MAS where agents are situated. Unfortunately, most previous research on value alignment of norms does not take into consideration the strong social dimension of values. Here, we take the stance that agents should be able to reason not exclusively about their own values, but also take into account the values that others in their community hold and how they interpret them. In this work, we present a novel functionality for autonomous agents to compute the perspective-dependent value alignment of norms. We build upon and integrate previous work on value representation, normative reasoning and Theory of Mind (the ability to perceive and interpret others in terms of their mental states, such as beliefs). This novel functionality enables an agent to compute the alignment of a set of norms with respect to a set of values not exclusively from its opinion perspective, but to switch its value structure and perception of the world at run-time using Theory of Mind, to estimate the alignment that another agent may have for the same set of norms. Our proposal opens new grounds for research on value-based negotiation over normative systems, where agents can perform better if they can estimate the opinion that their interlocutors have on the proposals they make.

Theory of Mind capabilities, i.e. the capacity to adopt and reason from the perspective of others. By combining the Theory of Mind of TomAbd agents with abductive reasoning, agents can infer explanations for the behaviour of others, which they can incorporate into their own decision-making. We have implemented the TomAbd agent model and successfully tested its performance in the cooperative board game Hanabi.

The deceleration of global poverty reduction in the last decades suggests that traditional redistribution policies are losing their effectiveness. Alternative ways to work towards the #1 United Nations Sustainable Development Goal (poverty eradication) are required. NGOs have insistingly denounced the criminalization of poverty, and the social science literature suggests that discrimination against the poor (a phenomenon known as aporophobia) could constitute a brake to the fight against poverty. This paper describes a proposal for an agent-based model to examine the impact that aporophobia at the institutional level has on poverty levels. This aporophobia agent-based model (AABM) will first be applied to a case study in the city of Barcelona. The regulatory environment is central to the model, since aporophobia has been identified in the legal framework. The AABM presented in this paper constitutes a cornerstone to obtain empirical evidence, in a non-invasive way, on the causal relationship between aporophobia and poverty levels. The simulations that will be generated based on the AABM have the potential to inform a new generation of poverty reduction policies, which act not only on the redistribution of wealth but also on the discrimination of the poor.

Ostrom's Institutional Analysis and Development (IAD) framework represents a comprehensive theoretical effort to identify and outline the variables that determine the outcome in any social interaction. Taking inspiration from it, we define the Action Situation Language (ASL), a machine-readable logical language to express the components of a multiagent interaction, with a special focus on the rules adopted by the community. The ASL is complemented by a game engine that takes an interaction description as input and automatically grounds its semantics as an Extensive-Form Game (EFG), which can be readily analysed using standard game-theoretical solution concepts. Overall, our model allows a community of agents to perform what-if analysis on a set of rules being considered for adoption, by automatically connecting rule configurations to the outcomes they incentivize.

Machine Ethics (ME) is concerned with the design of Artificial Moral Agents (AMAs), i.e. autonomous agents capable of reasoning and behaving according to moral values. Previous approaches have treated values as labels associated with some actions or states of the world, rather than as integral components of agent reasoning. It is also common to disregard that a value-guided agent operates alongside other value-guided agents in an environment governed by norms, thus omitting the social dimension of AMAs. In this blue sky paper, we propose a new AMA paradigm grounded in moral and social psychology, where values are instilled into agents as context-dependent goals. These goals intricately connect values at individual levels to norms at a collective level by evaluating the outcomes most incentivized by the norms in place. We argue that this type of normative reasoning, where agents are endowed with an understanding of norms' moral implications, leads to value-awareness in autonomous agents. Additionally, this capability paves the way for agents to align the norms enforced in their societies with respect to the human values instilled in them, by complementing the value-based reasoning on norms with agreement mechanisms to help agents collectively agree on the best set of norms that suit their human values. Overall, our agent model goes beyond the treatment of values as inert labels by connecting them to normative reasoning and to the social functionalities needed to integrate value-aware agents into our modern hybrid human-computer societies.

This paper presents a novel model, called T om A bd , that endows autonomous agents with Theory of Mind capabilities. T om A bd agents are able to simulate the perspective of the world that their peers have and reason from their perspective. Furthermore, T om A bd agents can reason from the perspective of others down to an arbitrary level of recursion , using Theory of Mind of n^th order. By combining the previous capability with abductive reasoning, T om A bd agents can infer the beliefs that others were relying upon to select their actions, hence putting them in a more informed position when it comes to their own decision-making. We have tested the T om A bd model in the challenging domain of Hanabi, a game characterised by cooperation and imperfect information. Our results show that the abilities granted by the T om A bd model boost the performance of the team along a variety of metrics, including final score, efficiency of communication, and uncertainty reduction.

The value alignment problem is concerned with the design of systems that provably abide by our human values. One approach to this challenge is through the leverage of prescriptive norms that, if carefully designed, are able to steer a multiagent system away from harmful outcomes and towards more beneficial ones. In this work, we first present a general methodology for the automated synthesis of value aligned normative systems, based on a consequentialist view of values. In the second part, we provide analytical tools to examine such value aligned normative systems, namely the Shapley value of individual norms and the compatibility of several values under a fixed set of norms. We illustrate all of our contributions with a running example of a society of agents where taxes are collected and redistributed according to a set of parametrised norms.

In this paper, we formalise and implement an agent model for cooperation under imperfect information. It is based on Theory of Mind (the cognitive ability to understand the mental state of others) and abductive reasoning (the inference paradigm that computes explanations from observations). The combination of these two techniques allows agents to derive the motives behind the actions of their peers, and incorporate this knowledge into their own decision-making. We have implemented this model in a totally domain-independent fashion and successfully tested it for the cooperative card game Hanabi.

This doctoral thesis is concerned with the engineering of values with an explicit pro-social (as opposed to a personal) focus. To do so, two approaches are explored, each dealing with a different level at which interactions are studied and engineered in a multi-agent system. The first, referred to as the collective approach, leverages prescriptive norms as the promoting mechanisms of pro-social values. The second, referred to as the individual approach, deals with the internal reasoning scheme of agents and endows them with the ability to reason about others. This results in empathetic autonomous agents, who are able to take the perspective of a peer and understand the motivations behind their behaviour.

Policies that seek to mitigate poverty by acting on equal opportunity have been found to aggravate discrimination against the poor (aporophobia), since individuals are made responsible for not progressing in the social hierarchy. Only a minority of the poor benefit from meritocracy in this era of growing inequality, generating resentment among those who seek to escape their needy situations by trying to climb up the ladder. Through the formulation and development of an agent-based social simulation, this study aims to analyse the role of norms implementing equal opportunity and social solidarity principles as enhancers or mitigators of aporophobia, as well as the threshold of aporophobia that would facilitate the success of poverty-reduction policies. The ultimate goal of the social simulation is to extract insights that could help inform and guide a new generation of policy making for poverty reduction by acting on the discrimination against the poor, in line with the UN “Leave No One Behind” principle. An “aporophobia-meter” will be developed and guidelines will be drafted based on both the simulation results and a review of poverty reduction policies at regional levels.

Value alignment has emerged in recent years as a basic principle to produce beneficial and mindful Artificial Intelligence systems. It mainly states that autonomous entities should behave in a way that is aligned with our human values. In this work, we summarize a previously developed model that considers values as preferences over states of the world and defines alignment between the governing norms and the values. We provide a use-case for this framework with the Iterated Prisoner’s Dilemma model, which we use to exemplify the definitions we review. We take advantage of this use-case to introduce new concepts to be integrated with the established framework: alignment equilibrium and Pareto optimal alignment. These are inspired on the classical Nash equilibrium and Pareto optimality, but are designed to account for any value we wish to model in the system.

The presented work concerns the study of solution sheared organic thin film transistors based on a 2,8-difluoro-5,11-bis(triethylsilylethynyl)anthradithiophene (diF-TES-ADT) polymer blend.

In recent years there has been an increasing interest in ensuring that autonomous systems behave consistently with human values. A popular approach to this challenge is through the incorporation of norms that regulate behaviour in an ethical way. However, such norms must be effective at promoting the values we consider most important. In this work, we introduce a systematic methodology for the automated synthesis of parametric normative systems based on value promotion. We introduce the new concepts of Shapley values of norms and value compatibility. To quantify the effectiveness of norms at upholding the values we consider relevant, we adopt the value alignment indicator from a previously established framework. We apply our model to a toy system which we use to illustrate our approach from end to end.

In the field of normative multiagent systems, the relationship between a game structure and its underpinning agent interaction rules is hardly ever addressed in a systematic manner. In this work, we introduce the Action Situation Language (ASL), inspired by Elinor Ostrom’s Institutional Analysis and Development framework, to bridge the gap between games and rules. The ASL provides a syntax for the description of agent interactions, and is complemented by an engine that automatically provides semantics for them as extensive-form games. The resulting games can then be analysed using standard game-theoretical solution concepts, hence allowing any community of agents to automatically perform what-if analysis of potential new interaction rules.