Stuart Russell defines the value alignment problem as follows: How can we build autonomous systems with values that “are aligned with those of the human race”? In this article I outline some distinctions that are useful for understanding the value alignment problem and then propose a solution: I argue that the methods currently applied by computer scientists for embedding moral values in autonomous systems can be improved by representing moral principles as conceptual spaces, i.e. as Voronoi (...) tessellations of morally similar choice situations located in a multidimensional geometric space. The advantage of my preferred geometric approach is that it can be implemented without specifying any utility function ex ante. (shrink)

How do we ensure that future generally intelligent AI share our values? This is the value-alignment problem. It is a weighty matter. After all, if AI are neutral with respect to our wellbeing, or worse, actively hostile toward us, then they pose an existential threat to humanity. Some philosophers have argued that one important way in which we can mitigate this threat is to develop only AI that shares our values or that has values that ‘align with’ ours. (...) However, there is nothing to guarantee that this policy will be universally implemented—in particular, ‘bad actors’ are likely to flout it. In this paper, I show how the predictive processing model of the mind, currently ascendant in cognitive science, may ameliorate the value-alignment problem. In essence, I argue that there are a plurality of reasons why any future generally intelligent AI will possess a predictive processing cognitive architecture (e.g. because we decide to build them that way; because it is the only possible cognitive architecture that can underpin general intelligence; because it is the easiest way to create AI.). I also argue that if future generally intelligent AI possess a predictive processing cognitive architecture, then they will come to share our pro-moral motivations (of valuing humanity as an end; avoiding maleficent actions; etc.), regardless of their initial motivation set. Consequently, these AI will pose a minimal threat to humanity. In this way then, I conclude, the value-alignment problem is significantly ameliorated under the assumption that future generally intelligent AI will possess a predictive processing cognitive architecture. (shrink)

As the capabilities of artificial intelligence systems improve, it becomes important to constrain their actions to ensure their behaviour remains beneficial to humanity. A variety of ethical, legal and safety-based frameworks have been proposed as a basis for designing these constraints. Despite their variations, these frameworks share the common characteristic that decision-making must consider multiple potentially conflicting factors. We demonstrate that these alignment frameworks can be represented as utility functions, but that the widely used Maximum Expected Utility paradigm provides (...) insufficient support for such multiobjective decision-making. We show that a Multiobjective Maximum Expected Utility paradigm based on the combination of vector utilities and non-linear action–selection can overcome many of the issues which limit MEU’s effectiveness in implementing aligned AI. We examine existing approaches to multiobjective AI, and identify how these can contribute to the development of human-aligned intelligent agents. (shrink)

Cocreating meaning in collaboration is challenging. Success is often determined by people's abilities to coordinate their language to converge upon shared mental representations. Here we explore one set of low‐level linguistic behaviors, linguistic alignment, that both emerges from, and facilitates, outcomes of high‐level convergence. Linguistic alignment captures the ways people reuse, that is, “align to,” the lexical, syntactic, and semantic forms of others' utterances. Our focus is on the temporal change of multi‐level linguistic alignment, as well as (...) how alignment is related to communicative outcomes within a unique collaborative problem‐solving paradigm. The primary task, situated within a virtual educational video game, requires creative thinking between three people where the paths for possible solutions are highly variable. We find that over time interactions are marked by decreasing lexical and syntactic alignment, with a trade‐off of increasing semantic alignment. However, greater semantic alignment does not translate into better team performance. Overall, these findings provide greater clarity on the role of linguistic coordination within complex and dynamic collaborative problem‐solving tasks. (shrink)

Tae Wan Kim, John Hooker, and Thomas Donaldson make an attempt, in recent articles, to solve the alignment problem. As they define the alignment problem, it is the issue of how to give AI systems moral intelligence. They contend that one might program machines with a version of Kantian ethics cast in deontic modal logic. On their view, machines can be aligned with human values if such machines obey principles of universalization and autonomy, as well as (...) a deontic utilitarian principle. Programming machines to do so might be useful, in their view, for applications such as future autonomous vehicles. Their proposal draws both on traditional logic-based and contemporary connectionist approaches, to fuse factual information with normative principles. I will argue that this approach makes demands of machines that go beyond what is currently feasible, and may extend past the limits of the possible for AI. I also argue that a deontological ethics for machines should place greater stress on the formula of humanity of the Kantian categorical imperative. On this principle, one ought never treat a person as a mere means. Recognition of what makes a person a person requires ethical insight. Similar insight is needed to tell treatment as a means from treatment as a mere means. The resources in Kim, Hooker, and Donaldson’s approach is insufficient for this reason. Hesitation regarding deployment of autonomous machines is warranted in light of these alignment concerns. (shrink)

Lexical alignment, a communication phenomenon where conversational partners adapt their word choices to become more similar, plays an important role in the development of language and social communication skills. While this has been studied extensively in the conversations of preschool‐aged children and their parents in Western, Educated, Industrialized, Rich, and Democratic (WEIRD) communities, research in other pediatric populations is sparse. This study makes significant expansions on the existing literature by focusing on alignment in naturalistic conversations of school‐aged children (...) from a non‐WEIRD population across multiple conversational tasks and with different types of adult partners. Typically developing children aged 5 to 8 years (n = 45) engaged in four semi‐structured conversations that differed by task (problem‐solving vs. play‐based) and by partner (parent vs. university student), resulting in a corpus of 180 conversations. Lexical alignment scores were calculated and compared to sham conversations, representing alignment occurring at the level of chance. Both children and adults coordinated their conversational utterances by re‐using or aligning each other's word choices. This alignment behavior persisted across conversational tasks and partners, although the degree of alignment was moderated by the conversational context. These findings suggest that lexical alignment is a robust phenomenon in conversations between school‐age children and adults. Furthermore, this study extends lexical alignment findings to a non‐WEIRD culture, suggesting that alignment may be a coordination strategy employed by adults and children across diverse linguistic and cultural backgrounds. (shrink)

This paper looks at philosophical questions that arise in the context of AI alignment. It defends three propositions. First, normative and technical aspects of the AI alignment problem are interrelated, creating space for productive engagement between people working in both domains. Second, it is important to be clear about the goal of alignment. There are significant differences between AI that aligns with instructions, intentions, revealed preferences, ideal preferences, interests and values. A principle-based approach to AI (...) class='Hi'>alignment, which combines these elements in a systematic way, has considerable advantages in this context. Third, the central challenge for theorists is not to identify ‘true’ moral principles for AI; rather, it is to identify fair principles for alignment that receive reflective endorsement despite widespread variation in people’s moral beliefs. The final part of the paper explores three ways in which fair principles for AI alignment could potentially be identified. (shrink)

Human beings are internally inconsistent in various ways. One way to develop this thought involves using the language of value alignment: the values we hold are not always aligned with our behavior, and are not always aligned with each other. Because of this self-misalignment, there is room for potential projects of human enhancement that involve achieving a greater degree of value alignment than we presently have. Relatedly, discussions of AI ethics sometimes focus on what is known as the (...) value alignment problem, the challenge of how to build AI that acts in accordance with our human values. We argue that there is an especially close connection between solving the value alignment problem in AI ethics and using AI to pursue certain forms of human enhancement. But in addition, we also argue that there are important limits to what kinds of human enhancement can be pursued in this way, because some forms of human enhancement—namely moral revolutions—involve a kind of value misalignment rather than alignment. (shrink)

The “Value Alignment Problem” is the challenge of how to align the values of artificial intelligence with human values, whatever they may be, such that AI does not pose a risk to the existence of humans. Existing approaches appear to conceive of the problem as "how do we ensure that AI solves the problem in the right way", in order to avoid the possibility of AI turning humans into paperclips in order to “make more paperclips” or (...) eradicating the human race to “solve climate change”. This paper proposes an approach to Alignment rooted in the Enactive theory of mind that reconceptualises it as "how do we make relevant to AI what it relevant to humans". This conceptualisation is supported with a discussion of 4E cognition and goes on to suggest that the Alignment problem and the Frame problem are the same problem. The paper concludes with a discussion of the tradeoffs of the different approaches to value alignment. (shrink)

The AGI alignment problem has a bimodal distribution of outcomes with most outcomes clustering around the poles of total success and existential, catastrophic failure. Consequently, attempts to solve AGI alignment should, all else equal, prefer false negatives (ignoring research programs that would have been successful) to false positives (pursuing research programs that will unexpectedly fail). Thus, we propose adopting a policy of responding to points of philosophical and practical uncertainty associated with the alignment problem by (...) limiting and choosing necessary assumptions to reduce the risk of false positives. Herein we explore in detail two relevant points of uncertainty that AGI alignment research hinges on---meta-ethical uncertainty and uncertainty about mental phenomena---and show how to reduce false positives in response to them. (shrink)

The “Value Alignment Problem” is the challenge of how to align the values of artificial intelligence with human values, whatever they may be, such that AI does not pose a risk to the existence of humans. A fundamental feature of how the problem is currently understood is that AI systems do not take the same things to be relevant as humans, whether turning humans into paperclips in order to “make more paperclips” or eradicating the human race to (...) “solve climate change”. Specifically, existing approaches approaches to alignment appear to be concerned with how AI might *solve* problems in the relevant way. This paper presents and explores an approach to alignment rooted in the Enactive Theory of mind. It offers an alternative conception of the alignment as “how do we make relevant to AI what is relevant to humans?” In this conception, alignment is concerned with building AI so that it *discerns* and defines the problem in the relevant way. In this way, the Alignment Problem is shown to be the same problem as the Frame Problem. The paper concludes with a consideration of tradeoffs between these conceptions of the alignment problem. (shrink)

As artificial intelligence systems are becoming increasingly autonomous and will soon be able to make decisions on their own about what to do, AI researchers have started to talk about the need to align AI with human values. The AI ‘value alignment problem’ faces two kinds of challenges—a technical and a normative one—which are interrelated. The technical challenge deals with the question of how to encode human values in artificial intelligence. The normative challenge is associated with two questions: (...) “Which values or whose values should artificial intelligence align with?” My concern is that AI developers underestimate the difficulty of answering the normative question. They hope that we can easily identify the purposes we really desire and that they can focus on the design of those objectives. But how are we to decide which objectives or values to induce in AI, given that there is a plurality of values and moral principles and that our everyday life is full of moral disagreements? In my paper I will show that although it is not realistic to reach an agreement on what we, humans, really want as people value different things and seek different ends, it may be possible to agree on what we do not want to happen, considering the possibility that intelligence, equal to our own, or even exceeding it, can be created. I will argue for pluralism which is compatible with objectivism. In spite of the fact that there is no uniquely best solution to every moral problem, it is still possible to identify which answers are wrong. And this is where we should begin the value alignment of AI. (shrink)

The alignment problem—establishing links between corresponding phrases in two related sentences—is as important in natural language inference (NLI) as it is in machine translation (MT). But the tools and techniques of MT alignment do not readily transfer to NLI, where one cannot assume semantic equivalence, and for which large volumes of bitext are lacking. We present a new NLI aligner, the MANLI system, designed to address these challenges. It uses a phrase-based alignment representation, exploits external lexical (...) resources, and capitalizes on a new set of supervised training data. We compare the performance of MANLI to existing NLI and MT aligners on an NLI alignment task over the well-known Recognizing Textual Entailment data. We show that MANLI significantly outperforms existing aligners, achieving gains of 6.2% in F1 over a representative NLI aligner and 10.5% over GIZA++. (shrink)

Solutions to word problems are moderated by the semantic alignment of real-world relations with mathematical operations. Categorical relations between entities are aligned with addition, whereas certain functional relations between entities are aligned with division. Similarly, discreteness vs. continuity of quantities is aligned with different formats for rational numbers. These alignments have been found both in textbooks and in the performance of college students in the USA and in South Korea. The current study examined evidence for alignments in Russia. Textbook (...) analyses revealed semantic alignments for arithmetic word problems, but not for rational numbers. Nonetheless, Russian college students showed semantic alignments both for arithmetic operations and for rational numbers. Since Russian students exhibit semantic alignments for rational numbers in the absence of exposure to examples in school, such alignments likely reflect intuitive understanding of mathematical representations of real-world situations. (shrink)

Entity alignment is the task of integrating heterogeneous knowledge among different knowledge graphs (KGs). KG is a popular way of storing facts about real-world entities. Unfortunately, a very limited number of the entities stored in different KGs are aligned. This paper presents an embedding-based entity alignment method that finds entity alignment by measuring the similarities between entity embeddings. Existing methods mainly focus on the relational structures and attributes information for the alignment process. Such methods fail while (...) the entities have a fewer number of attributes or when the relational structure could not capture the meaningful representation of the entities. To address this problem, we propose EASAE, an entity alignment method using summary and attribute embeddings. We exploit the entity summary information available in KGs for entities’ summary embedding. To learn the semantics of the entity summary, we employ bidirectional encoder representations from transformers (BERT). Our model learns the representations of entities by using relational triples, attribute triples and summary as well. We perform experiments on real-world datasets, and the results indicate that the proposed approach outperformed the state-of-the-art models for entity alignment. (shrink)

IN “CONSTRUCTIVISM, AGENCY, AND THE PROBLEM of Alignment,” Michael Bratman considers how lessons from the philosophy of action bear on the question of how best to construe the agent’s standpoint in the context of a constructivist theory of practical reasons. His focus is “the problem of alignment”: “whether the pressures from the general constructivism will align with the pressures from the theory of agency” (Bratman 2012: 81). He thus brings two lively literatures into dialogue with each (...) other. This is laudable. However, I shall argue that the considerations Bratman brings to bear from the literature on action do not support his conclusion that evaluative judgments are not constitutive of the agent’s standpoint in the context of constructivism. (shrink)

Correction of sensory transmission delays is an intractable problem, since there is no absolute reference for calibration. Phase-alignment is a practical alternative solution and can be realized by adaptive filters that operate locally with simple error signals.

Artificial Intelligence (AI) must be directed at humane ends. The development of AI has produced great uncertainties of ensuring AI alignment with human values (AI value alignment) through AI operations from design to use. For the purposes of addressing this problem, we adopt the phenomenological theories of material values and technological mediation to be that beginning step. In this paper, we first discuss the AI value alignment from the relevant AI studies. Second, we briefly present what (...) are material values and technological mediation and reflect on the AI value alignment through the lenses of these theories. We conclude that a set of finite human values can be defined and adapted to the stable life tasks that AI systems will be called upon to accomplish. The AI value alignment can also be fostered between designers and users through technological mediation. Upon that foundation, we propose a set of common principles to understand the AI value alignment through phenomenological theories. This paper contributes the unique knowledge of phenomenological theories to the discourse on AI alignment with human values. (shrink)

AI research is being challenged with ensuring that autonomous agents learn to behave ethically, namely in alignment with moral values. Here, we propose a novel way of tackling the value alignment problem as a two-step process. The first step consists on formalising moral values and value aligned behaviour based on philosophical foundations. Our formalisation is compatible with the framework of (Multi-Objective) Reinforcement Learning, to ease the handling of an agent’s individual and ethical objectives. The second step consists (...) in designing an environment wherein an agent learns to behave ethically while pursuing its individual objective. We leverage on our theoretical results to introduce an algorithm that automates our two-step approach. In the cases where value-aligned behaviour is possible, our algorithm produces a learning environment for the agent wherein it will learn a value-aligned behaviour. (shrink)

Using data from American emergency call centers, this article focuses on the coordination, and mutual relevance, of participants’ effort to manage two forms of unit completion – sequence closing and concluding the occasion in which the project was pursued. In doing so, we specify the import of sequence organization as one method for conducting, organizing, and resolving interactional projects participants may be said to pursue, and describe a range of possible relations between project completion and occasion closure and the locations (...) from which problems come to be introduced as parties move to resolve projects and close calls. As we show, sequence and occasion closings produced in the service of projects are fateful: they inexorably demand that the participants arrive at some alignment – or make visible their failure to do so – regarding the projects pursued in it, the status of those projects, and thus who, as a consequence, the parties are for another, that is, their ‘identities’. For strangers and familiars both, the management of projects and the manner in which closing is achieved matters. (shrink)

This paper presents a new trilemma with respect to resolving the control and alignment problems in machine ethics. Section 1 outlines three possible types of artificial moral agents (AMAs): (1) 'Inhuman AMAs' programmed to learn or execute moral rules or principles without understanding them in anything like the way that we do; (2) 'Better-Human AMAs' programmed to learn, execute, and understand moral rules or principles somewhat like we do, but correcting for various sources of human moral error; and (3) (...) 'Human-Like AMAs' programmed to understand and apply moral values in broadly the same way that we do, with a human-like moral psychology. Sections 2–4 then argue that each type of AMA generates unique control and alignment problems that have not been fully appreciated. Section 2 argues that Inhuman AMAs are likely to behave in inhumane ways that pose serious existential risks. Section 3 then contends that Better-Human AMAs run a serious risk of magnifying some sources of human moral error by reducing or eliminating others. Section 4 then argues that Human-Like AMAs would not only likely reproduce human moral failures, but also plausibly be highly intelligent, conscious beings with interests and wills of their own who should therefore be entitled to similar moral rights and freedoms as us. This generates what I call the New Control Problem: ensuring that humans and Human-Like AMAs exert a morally appropriate amount of control over each other. Finally, Section 5 argues that resolving the New Control Problem would, at a minimum, plausibly require ensuring what Hume and Rawls term ‘circumstances of justice’ between humans and Human-Like AMAs. But, I argue, there are grounds for thinking this will be profoundly difficult to achieve. I thus conclude on a skeptical note. Different approaches to developing ‘safe, ethical AI’ generate subtly different control and alignment problems that we do not currently know how to adequately resolve, and which may or may not be ultimately surmountable. (shrink)

Providing meaningful and actionable explanations for end-users is a situated problem requiring the intersection of multiple disciplines to address social, operational, and technical challenges. However, the explainable artificial intelligence community has not commonly adopted or created tangible design tools that allow interdisciplinary work to develop reliable AI-powered solutions. This paper proposes a formative architecture that defines the explanation space from a user-inspired perspective. The architecture comprises five intertwined components to outline explanation requirements for a task: (1) the end-users’ mental (...) models, (2) the end-users’ cognitive process, (3) the user interface, (4) the Human-Explainer Agent, and (5) the agent process. We first define each component of the architecture. Then, we present the Abstracted Explanation Space, a modeling tool that aggregates the architecture’s components to support designers in systematically aligning explanations with end-users’ work practices, needs, and goals. It guides the specifications of what needs to be explained (content: end-users’ mental model), why this explanation is necessary (context: end-users’ cognitive process), to delimit how to explain it (format: Human-Explainer Agent and user interface), and when the explanations should be given. We then exemplify the tool’s use in an ongoing case study in the aircraft maintenance domain. Finally, we discuss possible contributions of the tool, known limitations or areas for improvement, and future work to be done. (shrink)

Data heterogeneity is the obstacle for the resource sharing on Semantic Web, and ontology is regarded as a solution to this problem. However, since different ontologies are constructed and maintained independently, there also exists the heterogeneity problem between ontologies. Ontology matching is able to identify the semantic correspondences of entities in different ontologies, which is an effective method to address the ontology heterogeneity problem. Due to huge memory consumption and long runtime, the performance of the existing ontology (...) matching techniques requires further improvement. In this work, an extended compact genetic algorithm-based ontology entity matching technique is proposed, which uses both the compact encoding mechanism and linkage learning approach to match the ontologies efficiently. Compact encoding mechanism does not need to store and maintain the whole population in the memory during the evolving process, and the utilization of linkage learning protects the chromosome’s building blocks, which is able to reduce the algorithm’s running time and ensure the alignment’s quality. In the experiment, ECGA-OEM is compared with the participants of ontology alignment evaluation initiative and the state-of-the-art ontology matching techniques, and the experimental results show that ECGA-OEM is both effective and efficient. (shrink)

While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learning approach that is tailored to the ontology matching task. Our approach is based on embedding ontological terms in a high-dimensional Euclidean space. This embedding is derived on the basis of a novel phrase retrofitting strategy through which (...) semantic similarity information becomes inscribed onto fields of pre-trained word vectors. The resulting framework also incorporates a novel outlier detection mechanism based on a denoising autoencoder that is shown to improve performance. An ontology matching system derived using the proposed framework achieved an F-score of 94% on an alignment scenario involving the Adult Mouse Anatomical Dictionary and the Foundational Model of Anatomy ontology (FMA) as targets. This compares favorably with the best performing systems on the Ontology Alignment Evaluation Initiative anatomy challenge. We performed additional experiments on aligning FMA to NCI Thesaurus and to SNOMED CT based on a reference alignment extracted from the UMLS Metathesaurus. Our system obtained overall F-scores of 93.2% and 89.2% for these experiments, thus achieving state-of-the-art results. (shrink)

Evidence suggests that the incidence of research misconduct is not in decline despite efforts to improve awareness, education and governance mechanisms. Two responses to this problem are favoured: first, the promotion of an agent-centred ethics approach to enhance researchers’ personal responsibility and accountability, and second, a change in research culture to relieve perceived pressures to engage in misconduct. This article discusses the challenges for both responses and explains how normative coherence through values alignment might assist. We argue that (...) research integrity and research ethics convey mixed messages, which are likely to contribute to a form of normative confusion. For the successful adoption of an agent-centred approach, normative coherence is needed between the two. To facilitate normative coherence, we propose that research ethics and research integrity be underpinned by a shared set of moral values that can be enacted via codes and guidelines and imbue research environments. Furthermore, to facilitate culture change, the same normative coherence is necessary at all levels of an institution. Only via values alignment between institutional aims, management, institutional practices and researchers can an ethical culture become truly embedded in research institutions. (shrink)

The problem of evil is the problem of reconciling the world's afflictions with the fundamental attributes and justice of God. Throughout their lives people encounter painful events originating from nature and other individuals. Furthermore, it is believed that God created everything, particularly in divine religions. Scholars and thinkers have debated for centuries why an omniscient, omnipotent, just, and compassionate God would create evil. The problem of evil is sometimes employed by atheists as evidence against religion, and at (...) times as philosophical proof for the absence of God. Nowadays, it exerts a profound impact on people's beliefs within social media platforms. This article aims to address doubts and objections surrounding the problem of evil. Numerous studies have been conducted on this current issue. The subject, which is basically examined as natural and moral evil in the philosophy of religion, has been examined under the titles such as justice and cruelty, goodness and evil, will and power in classical theology. In this study, the subject is examined in terms of the relationship between the subject of evil and its object, and attention is drawn to the role of man in the formation of evil. In this context, a general overview of the problem of evil is presented in the introduction. The second section delves into how the problem of evil is approached in philosophical and theological traditions, specifically examining the viewpoints of classical kalam schools. In the later parts of the study, the subject has been tried to be reinterpreted under different titles. Within this framework, it is suggested that some doubts and questions about the problem of evil arise from a lack of understanding of religion, while others stem from interpretations perceived as religious within the tradition. To propose a solution to the problem, the concept of God, known for its power and authority in classical times, should be reevaluated. In the context of Islam, God should be considered as an entity possessing complete knowledge, absolute will and power, and creating the universe for goodness, possesses justice and mercy and bestows numerous blessings upon humanity. This perspective finds ample support in the verses of the Qur'an. The problem of evil sometimes emerges from misconceiving the relationship between God and nature as equivalent to a God-human relationship. However, God designed the universe's order to support the emergence of life in nature and make it conducive to the lives of all living beings, particularly humans. Natural phenomena and events unfold by the laws set by Allah. Evils such as diseases, fires, earthquakes, floods, storms, volcanic eruptions, rain, and lightning are not events directly orchestrated by God to harm humans. The harm they cause is primarily a result of humans' negative interactions with nature. Precautions can mitigate these harms. Yet, humans who pollute air, water, and forests and build poorly located structures hastily attribute the resulting evils to God. A careful examination of the numerous blessings in nature underscores that the universe was created by God as a benevolent order. God created humanity to undergo tests within such a world, aiming for human happiness in both this world and the hereafter. Thus, the entirety of God's will be goodness. Evil, conversely, arises from humanity's exercise of free will and the potential to commit malevolent acts. Consequently, evil stems from human interactions with nature and fellow beings. Destiny is another frequently misconstrued aspect. It refers to the natural and moral laws established by Almighty Allah to govern the order of nature and human existence. Destiny presents a realm of possibilities and does not coerce anyone's actions. To address the problem of evil, the portrayal of God, whose sole intent is good, must be reconstructed. Additionally, concepts like testing, fate, death, and sustenance need to be reinterpreted in alignment with this conception. Furthermore, methods that facilitate the acquisition of accurate religious information through social media should be developed to suit contemporary requirements. (shrink)

When discussing AI alignment, we usually refer to the problem of teaching or training advanced autonomous AI systems to make decisions that are aligned with human values or preferences. Proponents of this approach believe it can be employed as means to stay in control over sophisticated intelligent systems, thus avoiding certain existential risks. We identify three general obstacles on the path to implementation of value alignment: a technological/technical obstacle, a normative obstacle, and a calibration problem. Presupposing, (...) for the purposes of this discussion, that the technical and normative problems are solved, we focus on the problem of how to calibrate a system, for a specific value, to be on a specific location within a spectrum stretching between righteous and normal or average human behavior. Calibration, or more specifically mis-calibration, also raises the issue of trustworthiness. If we cannot trust AI systems to perform tasks the way we intended, we would not use them on our roads and at our homes. In an era where we strive to construct autonomous machines endowed with common sense, reasoning abilities and a connection to the world, so they would be able to act in alignment with human values, such mis-calibrations can make the difference between trustworthy and untrustworthy systems. (shrink)

Nietzsche and Heidegger pose important philosophical questions to science and its technological projects. The resultant contributes to what may be called a continental philosophy of science and the author argues that only such a rigorously critical approach to the question of science permits a genuinely philosophical reflection on science. More than a thoughtful reflection on science, however, the heart of philosophy is also at stake in such reflections. The author defends that if Nietzsche proposes the resources of art to defend (...) us against truth and the deadly insights of tragic knowledge, then Nietzsche's more arresting claim is his equation of science and art, just as Heidegger aligns techne and poiesis. For Nietzsche, science and art draw upon the same creative powers and both science and art are directed to the purpose of life. /// O ponto de partida deste artigo é o reconhecimento de que Friedrich Nietzsche e Martin Heidegger são dois filósofos que colocam questões profundamente relevantes acerca da Ciência e dos seus projectos tecnológicos. Neste sentido, o plano da autora consiste em demonstrar a viabilidade e a importância do modo não-analítico, ou continental, de fazer Filosofia da Ciência, argumentando que uma reflexão genuinamente filosófica acerca da Ciência não se pode dispensar de um confronto com o modo crítico de fazer filosofia representado tanto por Nietzsche como por Heidegger. Para a autora, os pensamentos críticos destes pensadores acerca da Ciência são bem mais do que uma mera reflexão filosófica acerca da Ciência; na verdade, o que aqui está em causa é propriamente saber de que se trata quando a questão é a da própria Filosofia. Assim, e na medida em que Nietzsche propõe os recursos da arte para nos defender contra as invectivas da verdade e as intuições fatais do conhecimento trágico, a autora do artigo defende que o mais interessante na posição nietzscheana tem a ver com o seu modo de equacionar a relação entre Ciência e Arte, tal como Heidegger acabará por alinhar techne e poiesis. O fundo da questão está em que para Nietzsche tanto a Ciência como a Arte recorrem aos mesmos poderes criativos, para além de que a Ciência e a Arte estão orientadas para a defesa do mesmo propósito: a Vida. (shrink)

Norms have been widely enacted in human and agent societies to regulate individuals’ actions. However, although legislators may have ethics in mind when establishing norms, moral values are only sometimes explicitly considered. This paper advances the state of the art by providing a method for selecting the norms to enact within a society that best aligns with the moral values of such a society. Our approach to aligning norms and values is grounded in the ethics literature. Specifically, from the literature’s (...) study of the relations between norms, actions, and values, we formally define how actions and values relate through the so-called value judgment function and how norms and values relate through the so-called norm promotion function. We show that both functions provide the means to compute value alignment for a set of norms. Moreover, we detail how to cast our decision-making problem as an optimisation problem: finding the norms that maximise value alignment. We also show how to solve our problem using off-the-shelf optimisation tools. Finally, we illustrate our approach with a specific case study on the European Value Study. (shrink)

This paper offers three contributions to the burgeoning movements of AI for Social Good (AI4SG) and AI and the United Nations Sustainable Development Goals (SDGs). First, we introduce the SDG-Intense Evaluation framework (SDGIE) that aims to situate variegated automated/AI models in a larger ecosystem of computational approaches to advance the SDGs. To foster knowledge collaboration for solving complex social and environmental problems encompassed by the SDGs, the SDGIE framework details a benchmark structure of data-algorithm-output to effectively standardize AI approaches to (...) the SDGs. Second, as a specific instantiation of the SDGIE framework, the SDG Impact Intensity Model (SDGIIM) is theoretically and operationally established. SDGIIM embeds expert decision-making and SDG keyword banks in textual data processing to determine overall SDG “impact intensity.” Ideally, SDGIIM can be applied to textual data sets from any sector or discipline: academia, business, government, non-profit, civil society, etc. Third, the SDGIIM instantiation is applied to the specific domain of academic journal rating systems as a case study. Traditionally, academic journals have been evaluated on loosely conceived and empirically shaky notions of ‘quality.’ Aligned with the trend of AI4SG and broader calls to action, ‘impact’ is rapidly becoming the primary normative consideration for assessing academic journals. We hypothesize and demonstrate that SDGIIM is capable of producing evaluations aligned with experts’ expectations of SDG impact intensity; the consistent analysis and rating of textual data sets that embody the SDGs with varying degrees of meaning and, ultimately, promote positive impact on the actual material conditions of the world. (shrink)

The Gongsun Longzi is often considered the only extant work of the Classical Chinese “School of Names”, an early intellectual tradition mainly concerned with logic and the philosophy of language. The Gongsun Longzi is a heterogeneous collection of five chapters that include short treatises and largely fictive dialogues between an anonymous persuader and his opponent, which typically revolve around a paradoxical claim. Its value as a testimony to Early Chinese philosophy, however, is somewhat controversial due to the intricate textual history (...) of the text and our limited knowledge about its intellectual backgrounds. This volume gathers contributions by leading specialists in the fields of Classical Chinese philosophy, philology, logic, and linguistics. Besides an overview of the scholarly literature on the topic and a detailed account of the reception of the text throughout time, it presents fresh insights into philological and philosophical problems raised by the Gongsun Longzi and other closely-related texts equally attributed to the “School of Names”. (shrink)

Multiple sequence alignment is one of the major problems that can be encountered in the bioinformatics field. MSA consists in aligning a set of biological sequences to extract the similarities between them. Unfortunately, this problem has been shown to be NP-hard. In this article, a new algorithm was proposed to deal with this problem; it is based on a quantum-inspired cuckoo search algorithm. The other feature of the proposed approach is the use of a randomized progressive (...) class='Hi'>alignment method based on a hybrid global/local pairwise algorithm to construct the initial population. The results obtained by this hybridization are very encouraging and show the feasibility and effectiveness of the proposed solution. (shrink)

Abstract In Reflective Knowledge, Ernest Sosa continues his detailed and intriguing defense of his two level account of knowledge that recognizes both animal and reflective knowledge. The latter more impressive type of knowledge requires a coherent positive epistemic perspective defending the reliability of a source of belief. Viewing Sosa's discussion from the through the lens provided by R.M. Chisholm's treatments of the problem of the criterion, I worry that Sosa's approach is too far in the methodist direction. As a (...) result, it is in danger of being unable to allow that paradigm examples of certain beliefs are indeed certain, e.g., the beliefs normal adults form in simple arithmetic truths under normal circumstances. I urge an approach closer to particularism that grants more weight to such paradigm particular cases. I also suggest that such an approach might actually align well with Sosa's coherentist sympathies. (shrink)

In this commentary on Michael Slote’s paper “The Many Faces of Empathy,” I assess the ways in which his theory of empathy aligns with simulation theory, as well as the problems that he needs to address because of this. Overall, I present three problems that need to be addressed: How do we know that we have caught the other’s emotion and not merely reacted on our own; What exactly is it about the other’s emotion or attitude that I am mimicking (...) and catching; and Does empathy provide us with objective or subjective ethical knowledge? (shrink)

Two-boxers in Newcomb Problems face the question: Why aren't you rich? The essay argues that one-boxers have a false sense of advantage. They fail to align their credences during deliberation with the credences they will have when they act. This puts them in violation of the so-called Principle of Reflection, and it exposes them to a dynamic Dutch Book that will leach any gains they achieve away.

This article addresses intraorganizational pressures for organizational transformation toward more responsible business practices by exploring the role of internal activists. Building on the interactive framing perspective, I ask how internal activists develop a framing of their company’s responsibilities as they attempt to transform its business practices from the inside out. I explore this question in the context of a Danish pharmaceutical company’s responsibilities regarding the rising diabetes problem. Grounded in an inductive, interpretive analysis, I show how internal activists developed (...) a framing of the company’s responsibility over time and eventually instigated new ways of thinking about and doing business in their organization. I theorize the constitutive processes that strengthened frame alignment and allowed the internal activists to shape business practices. My study contributes to the literature on corporate social responsibility (CSR) communication by explaining how intraorganizational processes of meaning making may constitute more responsible business practices and by explicating the distinct role of internal activists as agents of organizational transformation. (shrink)

This article offers a synthesis of certain essential contributions from three revolutionary thinkers of our age, C. West Churchman, Pir Vilayat Khan, and Ilya Prigogine, each of whom recently departed this life. In the course of this article, common threads in the work of these pioneers related to problem solving and creativity will be explored.

The standard moral realist response to the charge of “queerness” has been a type of companions in guilt defense. This response asserts that moral properties are no more queer than other abstract properties and since most people accept these other abstract properties into their ontology then they ought to also accept moral properties. Moral realists contend that moral properties face no special problem. In this paper, I argue that moral properties do face a unique problem – the moral (...) realist1 is not able to ac- count for the resemblance between moral properties in ways that the meta- physical realist has accounted for it between non-moral abstract proper- ties. The moral realist shares similar theoretical motivations as the realist about universals (i.e. those who accept other non-moral abstract proper- ties), and they attempt to align themselves with the realist about universals when defending their view against the charge of queerness. To highlight the unique����������������������������������������������������������������� the problem of accounting for resemblance between universals which has been made explicit in the metaphysical debate concerning realism. I then apply this problem to the seemingly parallel metaethical position, moral realism. Lastly, I describe one prominent metaphysical response that has been offered and argue that the moral realist cannot consistently accept this response. Moral properties do face a unique problem when compared with other abstract properties – accounting for the resemblance between ��������������������������������������������������������������������. (shrink)

New mechanical philosophy posits that explanations in the life sciences involve the decomposition of a system into its entities and their respective activities and organization that are responsible for the explanandum phenomenon. This mechanistic account of explanation has proven problematic in its application to mathematical models, leading the mechanists to suggest different ways of aligning abstract models with the mechanist program. Initially, the discussion centered on whether the Hodgkin-Huxley model is explanatory. Network models provided another complication, as they apply to (...) a wide number of materially diverse systems. In this article, we examine the various attempts to integrate abstract models within the mechanist program, also presenting a further challenge: the Heimburg-Jackson model, which was introduced as an alternative to the Hodgkin-Huxley model. We argue that although the notion of abstraction as the omission of irrelevant mechanistic details appears to give a mechanistic solution for accommodating abstract models, this notion does not suit models whose epistemic strategy is not decompositional. As a result, the mechanist has to choose whether to dilute the mechanistic approach nearly beyond recognition or to claim that many, if not most, abstract theoretical models do not deliver mechanistic explanations, or qualify as explanatory at all. (shrink)

Evans’ 1968 ANALOGY system was the first computer model of analogy. This paper demonstrates that the structure mapping model of analogy, when combined with high‐level visual processing and qualitative representations, can solve the same kinds of geometric analogy problems as were solved by ANALOGY. Importantly, the bulk of the computations are not particular to the model of this task but are general purpose: We use our existing sketch understanding system, CogSketch, to compute visual structure that is used by our existing (...) analogical matcher, Structure Mapping Engine (SME). We show how SME can be used to facilitate high‐level visual matching, proposing a role for structural alignment in mental rotation. We show how second‐order analogies over differences computed via analogies between pictures provide a more elegant model of the geometric analogy task. We compare our model against human data on a set of problems, showing that the model aligns well with both the answers chosen by people and the reaction times required to choose the answers. (shrink)

New mechanical philosophy posits that explanations in the life sciences involve the decomposition of a system into its entities and their respective activities and organization that are responsible for the explanandum phenomenon. This mechanistic account of explanation has proven problematic in its application to mathematical models, leading the mechanists to suggest different ways of aligning abstract models with the mechanist program. Initially, the discussion centered on whether the Hodgkin-Huxley model is explanatory. Network models provided another complication, as they apply to (...) a wide number of materially diverse systems. In this article, we examine the various attempts to integrate abstract models within the mechanist program, also presenting a further challenge: the Heimburg-Jackson model, which was introduced as an alternative to the Hodgkin-Huxley model. We argue that although the notion of abstraction as the omission of irrelevant mechanistic details appears to give a mechanistic solution for accommodating abstract models, this notion does not suit models whose epistemic strategy is not decompositional. As a result, the mechanist has to choose whether to dilute the mechanistic approach nearly beyond recognition or to claim that many, if not most, abstract theoretical models do not deliver mechanistic explanations, or qualify as explanatory at all. (shrink)

This manuscript draws on the moral norms arising from the nuanced accounts of epistemic (in)justice and social identity in relational autonomy to normatively assess and articulate the ethical problems associated with using AI in patient care in light of the Black Box problem. The article also describes how black-boxed AI may be used within the healthcare system. The manuscript highlights what needs to happen to align AI with the moral norms it draws on. Deeper thinking – from other backgrounds (...) other than decolonial scholarship and relational autonomy – about the impact of AI on the human experience needs to be done to appreciate any other barriers that may exist. Future studies can take up this task. (shrink)

Newcomb’s problem involves a decision-maker faced with a choice and a predictor forecasting this choice. The agents’ interaction seems to generate a choice dilemma once the decision-maker seeks to apply two basic principles of rational choice theory : maximize expected utility ; adopt the dominant strategy. We review unsuccessful attempts at pacifying the dilemma by excluding Newcomb’s problem as an RCT-application, by restricting MEU and ADS, and by allowing for backward causation. A probability approach shows that Newcomb’s original (...) problem-formulation lacks causal information. This makes it impossible to specify the probability structure of Newcomb’s univocally. Once missing information is added, Newcomb’s problem and RCT re-align, thus explaining Newcomb’s problem as a seeming dilemma. Building on Wolpert and Benford :1637–1646, 2013), we supply additional details and offer a crucial correction to their formal proof. (shrink)

Legislating comprehensive sex education curricula has long been believed to be essential to aligning education about sex, sexuality and human relationships with the values of equality, inclusivity and autonomy. Defences of the need for ‘good’ sex education in public schools are contingent upon arguments about whose experiences ought to guide us in determining what sufficient alignment with such values might look like. The aim of this paper is to explore the assumptions underlying one prevailing norm in such defences: what (...) I call parental deference or the practice of heeding to the rights of parents in debates about sex education. The question at the heart of this paper is, then, who in our communities does this deference exclude? I begin with a brief consideration of the appeal of parental deference within theories of education in general, and sex education in particular, before problematising its normalisation through a consideration of the exclusions such deference creates. In the process, I aim to explore the limits of parental deference for addressing the intersection of sexuality, sex, gender and race with violence in curricula and argue instead that one way of mitigating the consequences of these limits is to take seriously what I call sex education's community problem: If schools are to serve communities, why are so many community members’ experiences excluded? (shrink)

According to dispositionalism, fundamental properties are dispositions—powers that don’t reduce to other properties, laws, or anything else. As dispositions manifest, natural regularities result, so this view appears to explain the uniformity of nature. However, in this paper I’ll argue that there are types of regularities that can’t be explained by dispositionalism. The basic idea is this. All accounts of fundamental dispositions endow properties with a certain sort of structure. This allows explanations of only those regularities that align with such structures. (...) In this paper, I identify a type of natural regularity that cannot fit dispositionalist structures and show why the possibility of such regularities is problematic. (shrink)

Machine learning algorithms are increasingly used to support decision-making in the exercise of public authority. Here, we argue that an important consideration has been overlooked in previous discussions: whether the use of ML undermines the democratic legitimacy of public institutions. From the perspective of democratic legitimacy, it is not enough that ML contributes to efficiency and accuracy in the exercise of public authority, which has so far been the focus in the scholarly literature engaging with these developments. According to one (...) influential theory, exercises of administrative and judicial authority are democratically legitimate if and only if administrative and judicial decisions serve the ends of the democratic law maker, are based on reasons that align with these ends and are accessible to the public. These requirements are not satisfied by decisions determined through ML since such decisions are determined by statistical operations that are opaque in several respects. However, not all ML-based decision support systems pose the same risk, and we argue that a considered judgment on the democratic legitimacy of ML in exercises of public authority need take the complexity of the issue into account. This paper outlines considerations that help guide the assessment of whether a ML undermines democratic legitimacy when used to support public decisions. We argue that two main considerations are pertinent to such normative assessment. The first is the extent to which ML is practiced as intended and the extent to which it replaces decisions that were previously accessible and based on reasons. The second is that uses of ML in exercises of public authority should be embedded in an institutional infrastructure that secures reason giving and accessibility. (shrink)

How can we reconcile the fact that in order to act virtuously we appear to need to refer to the concept of a free will, while, at the same time, there are convincing philosophical arguments (aligned with a contemporary scientific understanding of natural causation) discrediting any viable notion of an unconstrained or uncaused will? Taking its cue from this important question, this chapter will proceed along the following lines. First, I aim to substantiate the link between contemporary character education and (...) the concept of the free will so as to illustrate the interdependency between the two. Next, I will scrutinize the concept of a free will, raising some philosophical concerns about its validity in a contemporary educational context. This involves looking at the philosophical stakes involved in proposing a unique capacity to intervene with the causal order of nature. At this point, I will suggest that there is a way out of this conundrum, and I will continue by proposing a radically different understanding of the will, offered by the seventeenth-century rationalist Spinoza. The chapter will close by looking at some of the practical consequences of grounding contemporary character education in a Spinozistic conception of the will. (shrink)