Results for 'Value Alignment Problem'

991 found
Order:
  1. The Value Alignment Problem.Dan J. Bruiger - manuscript
    The Value Alignment Problem (VAP) presupposes that artificial general intelligence (AGI) is desirable and perhaps inevitable. As usually conceived, it is one side of the more general issue of mutual control between agonistic agents. To be fully autonomous, an AI must be an autopoietic system (an agent), with its own purposiveness. In the case of such systems, Bostrom’s orthogonality thesis is untrue. The VAP reflects the more general problem of interfering in complex systems, entraining the possibility (...)
    Download  
     
    Export citation  
     
    Bookmark  
  2. AI Alignment Problem: “Human Values” Don’T Actually Exist.Alexey Turchin - manuscript
    Abstract. The main current approach to the AI safety is AI alignment, that is, the creation of AI whose preferences are aligned with “human values.” Many AI safety researchers agree that the idea of “human values” as a constant, ordered sets of preferences is at least incomplete. However, the idea that “humans have values” underlies a lot of thinking in the field; it appears again and again, sometimes popping up as an uncritically accepted truth. Thus, it deserves a thorough (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  3.  7
    An Enactive Approach to Value Alignment in Artificial Intelligence: A Matter of Relevance.Michael Cannon - 2021 - In Vincent C. Müller (ed.), Philosophy and Theory of AI. Springer Cham. pp. 119-135.
    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” (...)
    Download  
     
    Export citation  
     
    Bookmark  
  4.  8
    An Enactive Approach to Value Alignment in Artificial Intelligence: A Matter of Relevance.Michael Cannon - 2022 - In Vincent C. Müller (ed.), Philosophy and Theory of Artificial Intelligence 2021. pp. 119-135.
    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 (...)
    Download  
     
    Export citation  
     
    Bookmark  
  5.  23
    The Ghost in the Machine has an American Accent: Value Conflict in GPT-3.Rebecca Johnson, Giada Pistilli, Natalia Menedez-Gonzalez, Leslye Denisse Dias Duran, Enrico Panai, Julija Kalpokiene & Donald Jay Bertulfo - manuscript
    The alignment problem in the context of large language models must consider the plurality of human values in our world. Whilst there are many resonant and overlapping values amongst the world’s cultures, there are also many conflicting, yet equally valid, values. It is important to observe which cultural values a model exhibits, particularly when there is a value conflict between input prompts and generated outputs. We discuss how the co- creation of language and cultural value impacts (...)
    Download  
     
    Export citation  
     
    Bookmark  
  6.  92
    Dynamic Cognition Applied to Value Learning in Artificial Intelligence.Nythamar De Oliveira & Nicholas Corrêa - 2021 - Aoristo - International Journal of Phenomenology, Hermeneutics and Metaphysics 4 (2):185-199.
    Experts in Artificial Intelligence (AI) development predict that advances in the dvelopment of intelligent systems and agents will reshape vital areas in our society. Nevertheless, if such an advance isn't done with prudence, it can result in negative outcomes for humanity. For this reason, several researchers in the area are trying to develop a robust, beneficial, and safe concept of artificial intelligence. Currently, several of the open problems in the field of AI research arise from the difficulty of avoiding unwanted (...)
    Download  
     
    Export citation  
     
    Bookmark  
  7. Attention to Values Helps Shape Convergence Research.Casey Helgeson, Robert E. Nicholas, Klaus Keller, Chris E. Forest & Nancy Tuana - 2022 - Climatic Change 170.
    Convergence research is driven by specific and compelling problems and requires deep integration across disciplines. The potential of convergence research is widely recognized, but questions remain about how to design, facilitate, and assess such research. Here we analyze a seven-year, twelve-million-dollar convergence project on sustainable climate risk management to answer two questions. First, what is the impact of a project-level emphasis on the values that motivate and tie convergence research to the compelling problems? Second, how does participation in convergence projects (...)
    Download  
     
    Export citation  
     
    Bookmark  
  8. How Does Artificial Intelligence Pose an Existential Risk?Karina Vold & Daniel R. Harris - forthcoming - In Carissa Véliz (ed.), Oxford Handbook of Digital Ethics.
    Alan Turing, one of the fathers of computing, warned that Artificial Intelligence (AI) could one day pose an existential risk to humanity. Today, recent advancements in the field AI have been accompanied by a renewed set of existential warnings. But what exactly constitutes an existential risk? And how exactly does AI pose such a threat? In this chapter we aim to answer these questions. In particular, we will critically explore three commonly cited reasons for thinking that AI poses an existential (...)
    Download  
     
    Export citation  
     
    Bookmark  
  9.  67
    Improve Alignment of Research Policy and Societal Values.Peter Novitzky, Michael J. Bernstein, Vincent Blok, Robert Braun, Tung Tung Chan, Wout Lamers, Anne Loeber, Ingeborg Meijer, Ralf Lindner & Erich Griessler - 2020 - Science 369 (6499):39-41.
    Historically, scientific and engineering expertise has been key in shaping research and innovation policies, with benefits presumed to accrue to society more broadly over time. But there is persistent and growing concern about whether and how ethical and societal values are integrated into R&I policies and governance, as we confront public disbelief in science and political suspicion toward evidence-based policy-making. Erosion of such a social contract with science limits the ability of democratic societies to deal with challenges presented by new, (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  10.  21
    How Values Shape the Machine Learning Opacity Problem.Emily Sullivan - 2022 - In Insa Lawler, Kareem Khalifa & Elay Shech (eds.), Scientific Understanding and Representation. Routledge. pp. 306-322.
    One of the main worries with machine learning model opacity is that we cannot know enough about how the model works to fully understand the decisions they make. But how much is model opacity really a problem? This chapter argues that the problem of machine learning model opacity is entangled with non-epistemic values. The chapter considers three different stages of the machine learning modeling process that corresponds to understanding phenomena: (i) model acceptance and linking the model to the (...)
    Download  
     
    Export citation  
     
    Bookmark  
  11. Shortcuts to Artificial Intelligence.Nello Cristianini - forthcoming - In Marcello Pelillo & Teresa Scantamburlo (eds.), Machines We Trust. MIT Press.
    The current paradigm of Artificial Intelligence emerged as the result of a series of cultural innovations, some technical and some social. Among them are apparently small design decisions, that led to a subtle reframing of the field’s original goals, and are by now accepted as standard. They correspond to technical shortcuts, aimed at bypassing problems that were otherwise too complicated or too expensive to solve, while still delivering a viable version of AI. Far from being a series of separate problems, (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  12. Modelos Dinâmicos Aplicados à Aprendizagem de Valores em Inteligência Artificial.Nicholas Kluge Corrêa & Nythamar De Oliveira - 2020 - Veritas – Revista de Filosofia da Pucrs 2 (65):1-15.
    Experts in Artificial Intelligence (AI) development predict that advances in the development of intelligent systems and agents will reshape vital areas in our society. Nevertheless, if such an advance is not made prudently and critically-reflexively, it can result in negative outcomes for humanity. For this reason, several researchers in the area have developed a robust, beneficial, and safe concept of AI for the preservation of humanity and the environment. Currently, several of the open problems in the field of AI research (...)
    Download  
     
    Export citation  
     
    Bookmark  
  13.  59
    Quantum of Wisdom.Brett Karlan & Colin Allen - forthcoming - In Greg Viggiano (ed.), Quantum Computing and AI: Social, Ethical, and Geo-Political Implications. Hoboken, NJ: Wiley-Blackwell. pp. 1-6.
    Practical quantum computing devices and their applications to AI in particular are presently mostly speculative. Nevertheless, questions about whether this future technology, if achieved, presents any special ethical issues are beginning to take shape. As with any novel technology, one can be reasonably confident that the challenges presented by "quantum AI" will be a mixture of something new and something old. Other commentators (Sevilla & Moreno 2019), have emphasized continuity, arguing that quantum computing does not substantially affect approaches to (...) alignment methods for AI, although they allow that further questions arise concerning governance and verification of quantum AI applications. In this brief paper, we turn our attention to the problem of identifying as-yet-unknown discontinuities that might result from quantum AI applications. Wise development, introduction, and use of any new technology depends on successfully anticipating new modes of failure for that technology. This requires rigorous efforts to break systems in protected sandboxes, and it must be conducted at all stages of technology design, development, and deployment. Such testing must also be informed by technical expertise but cannot be left solely to experts in the technology because of the history of failures to predict how non-experts will use or adapt to new technologies. This interplay between experts and non-experts may be particularly acute for quantum AI because quantum mechanics is notoriously difficult to understand. (As Richard Feynman quipped, "Anyone who claims to understand quantum mechanics is either lying or crazy.") We will discuss the extent to which the difficulties in understanding the physics underlying quantum computing challenges attempts to anticipate new failure modes that might be introduced in AI applications intended for unsupervised operation in the public sphere. (shrink)
    Download  
     
    Export citation  
     
    Bookmark  
  14. An Expected Value Approach to the Dual-Use Problem.Thomas Douglas - 2013 - In Brian Rappert & Michael Selgelid (eds.), On the Dual Uses of Science and Ethics: Principles, Practices, and Prospects. ANU Press.
    In this chapter I examine how expected-value theory might inform responses to what I call the dual-use problem. I begin by defining that problem. I then outline a procedure, which invokes expected-value theory, for tackling it. I first illustrate the procedure with the aid of a simplified schematic example of a dual-use problem, and then describe how it might also guide responses to more complex real-world cases. I outline some attractive features of the procedure. Finally, (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  15. On Value and Obligation in Practical Reason: Toward a Resolution of the Is–Ought Problem in the Thomistic Moral Tradition.William Matthew Diem - 2021 - Nova et Vetera 19 (2): 531-562.
    Within the Thomistic moral tradition, the is-ought gap is regularly treated as identical to the fact-value gap, and these two dichotomies are also regularly treated as being identical to Aristotle and Aquinas’s distinction between the practical and speculative intellect. The question whether (and if so, how) practical (‘ought’) knowledge derives from speculative (‘is’) knowledge has driven some of the fiercest disputes among the schools of Thomistic natural lawyers. I intend to show that both of these identifications are wrong and (...)
    Download  
     
    Export citation  
     
    Bookmark  
  16. In Conversation with Artificial Intelligence: Aligning Language Models with Human Values.Atoosa Kasirzadeh - forthcoming - Philosophy and Technology.
    Large-scale language technologies are increasingly used in various forms of communication with humans across different contexts. One particular use case for these technologies is conversational agents, which output natural language text in response to prompts and queries. This mode of engagement raises a number of social and ethical questions. For example, what does it mean to align conversational agents with human norms or values? Which norms or values should they be aligned with? And how can this be accomplished? In this (...)
    Download  
     
    Export citation  
     
    Bookmark  
  17.  85
    The Epistemic Value of Conscious Acquaintance: A Problem for Reductive Physicalism.Adam Pautz - manuscript
    We take it that conscious acquaintance has great epistemic value. I develop a new problem for reductive physicalism concerning the epistemic value of acquaintance. The problem concerns "multiple candidate cases". (This develops a theme of my paper *The Significance Argument for the Irreducibility of Consciousness", Philosophical Perspectives 2017.).
    Download  
     
    Export citation  
     
    Bookmark  
  18. Superintelligence as a Cause or Cure for Risks of Astronomical Suffering.Kaj Sotala & Lukas Gloor - 2017 - Informatica: An International Journal of Computing and Informatics 41 (4):389-400.
    Discussions about the possible consequences of creating superintelligence have included the possibility of existential risk, often understood mainly as the risk of human extinction. We argue that suffering risks (s-risks) , where an adverse outcome would bring about severe suffering on an astronomical scale, are risks of a comparable severity and probability as risks of extinction. Preventing them is the common interest of many different value systems. Furthermore, we argue that in the same way as superintelligent AI both contributes (...)
    Download  
     
    Export citation  
     
    Bookmark   8 citations  
  19.  53
    Aligning Patient’s Ideas of a Good Life with Medically Indicated Therapies in Geriatric Rehabilitation Using Smart Sensors.Cristian Timmermann, Frank Ursin, Christopher Predel & Florian Steger - 2021 - Sensors 21 (24):8479.
    New technologies such as smart sensors improve rehabilitation processes and thereby increase older adults’ capabilities to participate in social life, leading to direct physical and mental health benefits. Wearable smart sensors for home use have the additional advantage of monitoring day-to-day activities and thereby identifying rehabilitation progress and needs. However, identifying and selecting rehabilitation priorities is ethically challenging because physicians, therapists, and caregivers may impose their own personal values leading to paternalism. Therefore, we develop a discussion template consisting of a (...)
    Download  
     
    Export citation  
     
    Bookmark  
  20.  81
    Look Who’s Talking: Responsible Innovation, the Paradox of Dialogue and the Voice of the Other in Communication and Negotiation Processes.Vincent Blok - 2014 - Journal of Responsible Innovation 1 (2):171-190.
    In this article, we develop a concept of stakeholder dialogue in responsible innovation (RI) processes. The problem with most concepts of communication is that they rely on ideals of openness, alignment and harmony, even while these ideals are rarely realized in practice. Based on the work of Burke, Habermas, Deetz and Levinas, we develop a concept of stakeholder dialogue that is able to deal with fundamentally different interests and value frames of actors involved in RI processes. We (...)
    Download  
     
    Export citation  
     
    Bookmark   20 citations  
  21. On Pritchard, Objectual Understanding and the Value Problem.J. Adam Carter & Emma C. Gordon - forthcoming - American Philosophical Quarterly.
    Duncan Pritchard (2008, 2009, 2010, forthcoming) has argued for an elegant solution to what have been called the value problems for knowledge at the forefront of recent literature on epistemic value. As Pritchard sees it, these problems dissolve once it is recognized that that it is understanding-why, not knowledge, that bears the distinctive epistemic value often (mistakenly) attributed to knowledge. A key element of Pritchard’s revisionist argument is the claim that understanding-why always involves what he calls strong (...)
    Download  
     
    Export citation  
     
    Bookmark   3 citations  
  22. Knowledge as a Thick Concept: New Light on the Gettier and Value Problems.Brent G. Kyle - 2011 - Dissertation, Cornell University
    I argue that knowledge is a particular kind of concept known as a thick concept. Examples of thick concepts include courage, generosity, loyalty, brutality, and so forth. These concepts are commonly said to combine both evaluation and description, and one of the main goals of this dissertation is to provide a new account of how a thick concept combines these elements. It is argued that thick concepts are semantically evaluative, and that they combine evaluation and description in a way similar (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  23. Value Judgements and Value Neutrality in Economics.Philippe Mongin - 2006 - Economica 73 (290):257-286.
    The paper analyses economic evaluations by distinguishing evaluative statements from actual value judgments. From this basis, it compares four solutions to the value neutrality problem in economics. After rebutting the strong theses about neutrality (normative economics is illegitimate) and non-neutrality (the social sciences are value-impregnated), the paper settles the case between the weak neutrality thesis (common in welfare economics) and a novel, weak non-neutrality thesis that extends the realm of normative economics more widely than the other (...)
    Download  
     
    Export citation  
     
    Bookmark   15 citations  
  24. The Undetectable Difference: An Experimental Look at the ‘Problem’ of P-Values.William M. Goodman - 2010 - Statistical Literacy Website/Papers: Www.Statlit.Org/Pdf/2010GoodmanASA.Pdf.
    In the face of continuing assumptions by many scientists and journal editors that p-values provide a gold standard for inference, counter warnings are published periodically. But the core problem is not with p-values, per se. A finding that “p-value is less than α” could merely signal that a critical value has been exceeded. The question is why, when estimating a parameter, we provide a range (a confidence interval), but when testing a hypothesis about a parameter (e.g. µ (...)
    Download  
     
    Export citation  
     
    Bookmark  
  25.  49
    Numerical Methods for Solving Initial Value Problems on the Infinity Computer.Yaroslav Sergeyev, Marat Mukhametzhanov, Francesca Mazzia, Felice Iavernaro & Pierluigi Amodio - 2016 - International Journal of Unconventional Computing 12 (1):3-23.
    New algorithms for the numerical solution of Ordinary Differential Equations (ODEs) with initial condition are proposed. They are designed for work on a new kind of a supercomputer – the Infinity Computer, – that is able to deal numerically with finite, infinite and infinitesimal numbers. Due to this fact, the Infinity Computer allows one to calculate the exact derivatives of functions using infinitesimal values of the stepsize. As a consequence, the new methods described in this paper are able to work (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  26. Epistemic Value, Duty, and Virtue.Guy Axtell - forthcoming - In Brian C. Barnett (ed.), Introduction to Philosophy: Epistemology. Rebus Community.
    This chapter introduces some central issues in Epistemology, and, like others in the open textbook series Introduction to Philosophy, is set up for rewarding college classroom use, with discussion/reflection questions matched to clearly-stated learning objectives,, a brief glossary of the introduced/bolded terms/concepts, links to further open source readings as a next step, and a readily-accessible outline of the classic between William Clifford and William James over the "ethics of belief." The chapter introduces questions of epistemic value through Plato's famous (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  27. Aligning with the Good.Benjamin Mitchell-Yellin - 2015 - Journal of Ethics and Social Philosophy (2):1-8.
    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 (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  28. Robustness to Fundamental Uncertainty in AGI Alignment.G. G. Worley Iii - 2020 - Journal of Consciousness Studies 27 (1-2):225-241.
    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 (...)
    Download  
     
    Export citation  
     
    Bookmark  
  29. Just the Right Thickness: A Defense of Second-Wave Virtue Epistemology.Guy Axtell & J. Adam Carter - 2008 - Philosophical Papers 37 (3):413-434.
    Abstract Do the central aims of epistemology, like those of moral philosophy, require that we designate some important place for those concepts located between the thin-normative and the non-normative? Put another way, does epistemology need "thick" evaluative concepts and with what do they contrast? There are inveterate traditions in analytic epistemology which, having legitimized a certain way of viewing the nature and scope of epistemology's subject matter, give this question a negative verdict; further, they have carried with them a tacit (...)
    Download  
     
    Export citation  
     
    Bookmark   13 citations  
  30. Many-Valued Logics. A Mathematical and Computational Introduction.Luis M. Augusto - 2020 - London: College Publications.
    2nd edition. Many-valued logics are those logics that have more than the two classical truth values, to wit, true and false; in fact, they can have from three to infinitely many truth values. This property, together with truth-functionality, provides a powerful formalism to reason in settings where classical logic—as well as other non-classical logics—is of no avail. Indeed, originally motivated by philosophical concerns, these logics soon proved relevant for a plethora of applications ranging from switching theory to cognitive modeling, and (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  31. Value Pluralism.Ruth Chang - 2015 - In James Wright (ed.), International Encyclopedia of the Social and Behavioral Sciences (Second Edition). Elsevier. pp. 21-26.
    Value pluralism’ as traditionally understood is the metaphysical thesis that there are many values that cannot be ‘reduced’ to a single supervalue. While it is widely assumed that value pluralism is true, the case for value pluralism depends on resolution of a neglected question in value theory: how are values properly individuated? Value pluralism has been thought to be important in two main ways. If values are plural, any theory that relies on value monism, (...)
    Download  
     
    Export citation  
     
    Bookmark   4 citations  
  32.  91
    Uncertain Values: An Axiomatic Approach to Axiological Uncertainty.Stefan Riedener - 2021 - Berlin, Germany: De Gruyter.
    How ought you to evaluate your options if you're uncertain about what's fundamentally valuable? A prominent response is Expected Value Maximisation (EVM)—the view that under axiological uncertainty, an option is better than another if and only if it has the greater expected value across axiologies. But the expected value of an option depends on quantitative probability and value facts, and in particular on value comparisons across axiologies. We need to explain what it is for such (...)
    Download  
     
    Export citation  
     
    Bookmark  
  33. Epistemic Value in the Subpersonal Vale.J. Adam Carter & Robert D. Rupert - 2020 - Synthese 198 (10):9243-9272.
    A vexing problem in contemporary epistemology—one with origins in Plato’s Meno—concerns the value of knowledge, and in particular, whether and how the value of knowledge exceeds the value of mere true opinion. The recent literature is deeply divided on the matter of how best to address the problem. One point, however, remains unquestioned: that if a solution is to be found, it will be at the personal level, the level at which states of subjects or (...)
    Download  
     
    Export citation  
     
    Bookmark   5 citations  
  34. Biomedical Ontology Alignment: An Approach Based on Representation Learning.Prodromos Kolyvakis, Alexandros Kalousis, Barry Smith & Dimitris Kiritsis - 2018 - Journal of Biomedical Semantics 9 (21).
    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 (...)
    Download  
     
    Export citation  
     
    Bookmark  
  35. Value Disagreement and Two Aspects of Meaning.Erich Rast - 2017 - Croatian Journal of Philosophy 17 (51):399-430.
    The problem of value disagreement and contextualist, relativist and metalinguistic attempts of solving it are laid out. Although the metalinguistic account seems to be on the right track, it is argued that it does not sufficiently explain why and how disagreements about the meaning of evaluative terms are based on and can be decided by appeal to existing social practices. As a remedy, it is argued that original suggestions from Putnam's 'The Meaning of "Meaning"' ought to be taken (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  36. The Limits of Conventional Justification: Inductive Risk and Industry Bias Beyond Conventionalism.Miguel Ohnesorge - 2020 - Frontiers in Research Metric and Analytics 14.
    This article develops a constructive criticism of methodological conventionalism. Methodological conventionalism asserts that standards of inductive risk ought to be justified in virtue of their ability to facilitate coordination in a research community. On that view, industry bias occurs when conventional methodological standards are violated to foster industry preferences. The underlying account of scientific conventionality, however, is problematically incomplete. Conventions may be justified in virtue of their coordinative functions, but often qualify for posterior empirical criticism as research advances. Accordingly, industry (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  37.  60
    A Peripatetic Argument for the Intrinsic Value of Human Life: Alexander of Aphrodisias' Ethical Problems I.Javier Echeñique - 2021 - Apeiron: A Journal for Ancient Philosophy and Science 54 (3):367-384.
    In this article I argue for the thesis that Alexander's main argument, in Ethical Problems I, is an attempt to block the implication drawn by the Stoics and other ancient philosophers from the double potential of use exhibited by human life, a life that can be either well or badly lived. Alexander wants to resist the thought that this double potential of use allows the Stoics to infer that human life, in itself, or by its own nature, is neither good (...)
    Download  
     
    Export citation  
     
    Bookmark  
  38.  94
    Robustness to Fundamental Uncertainty in AGI Alignment.I. I. I. G. Gordon Worley - manuscript
    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 metaphysical and practical uncertainty associated with the alignment problem by (...)
    Download  
     
    Export citation  
     
    Bookmark  
  39. Machines Learning Values.Steve Petersen - 2020 - In S. Matthew Liao (ed.), Ethics of Artificial Intelligence. New York, USA: Oxford University Press.
    Whether it would take one decade or several centuries, many agree that it is possible to create a *superintelligence*---an artificial intelligence with a godlike ability to achieve its goals. And many who have reflected carefully on this fact agree that our best hope for a "friendly" superintelligence is to design it to *learn* values like ours, since our values are too complex to program or hardwire explicitly. But the value learning approach to AI safety faces three particularly philosophical puzzles: (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
  40. Infinite Value and the Best of All Possible Worlds.Nevin Climenhaga - 2018 - Philosophy and Phenomenological Research 97 (2):367-392.
    A common argument for atheism runs as follows: God would not create a world worse than other worlds he could have created instead. However, if God exists, he could have created a better world than this one. Therefore, God does not exist. In this paper I challenge the second premise of this argument. I argue that if God exists, our world will continue without end, with God continuing to create value-bearers, and sustaining and perfecting the value-bearers he has (...)
    Download  
     
    Export citation  
     
    Bookmark   10 citations  
  41.  91
    Value-First Accounts of Reasons and Fit.R. A. Rowland - forthcoming - In Fittingness. Oxford University Press.
    It is tempting to think that all of normativity, such as our reasons for action, what we ought to do, and the attitudes that it is fitting for us to have, derives from what is valuable. But value-first approaches to normativity have fallen out of favour as the virtues of reasons- and fittingness-first approaches to normativity have become clear. On these views, value is not explanatorily prior to reasons and fit; rather the value of things is understood (...)
    Download  
     
    Export citation  
     
    Bookmark  
  42. Epistemic Problems of Utilitarian Practical Reasoning.John Dilworth - 1998-9 - Proceedings of the Heraclitean Society 19.
    Utilitarian (U.) theories must be capable of being applied in practical reasoning, or they would have no value as a guide to rational conduct. However, I show that epistemic extensions to U. theories produce logical confusion. Basic questions about what one needs to know in order to apply a U. analysis embroil one in an infinite regress. And attempts to incrementally apply U. either are no help at all (leaving one entirely 'in the dark'), or in general constitute arbitrary (...)
    Download  
     
    Export citation  
     
    Bookmark  
  43. The Other and the Subject: On the Conditions of Possibility of the Problem of Values in the Humanities.Anton Froeyman - forthcoming - In Gertrudis Van De Vijver & Boris Demarest (eds.), Critical Reflections on Objectivity. Georg Olms Verlag.
    Download  
     
    Export citation  
     
    Bookmark  
  44.  64
    Value and Idiosyncratic Fitting Attitudes.Conor McHugh & Jonathan Way - forthcoming - In Christopher Howard & Richard Rowland (eds.), Fittingness. Oxford University Press.
    Norm-attitude accounts of value say that for something to be valuable is for there to be norms that support valuing that thing. For example, according to fitting-attitude accounts, something is of value if it is fitting to value, and according to buck-passing accounts, something is of value if the reasons support valuing it. Norm-attitude accounts face the partiality problem: in cases of partiality, what it is fitting to value, and what the reasons support valuing, (...)
    Download  
     
    Export citation  
     
    Bookmark  
  45. The Value and Normative Role of Knowledge.Julien Dutant - 2014 - Liber Amicorum Pascal Engel.
    Why does knowledge matter? Two answers have been influential in the recent literature. One is that it has value: knowledge is one of the goods. Another is that it plays a significant normative role: knowledge is the norm of action, belief, assertion, or the like. This paper discusses whether one can derive one of the claims from the other. That is, whether assuming the idea that knowledge has value — and some defensible general hypotheses about norms and values (...)
    Download  
     
    Export citation  
     
    Bookmark   2 citations  
  46. Incommensurable values.John Broome - 2000 - In Roger Crisp & Brad Hooker (eds.), Well-Being and Morality: Essays in Honour of James Griffin. Clarendon Press. pp. 21--38.
    Two options are incommensurate in value if neither is better than the other, and if a small improvement or worsening of one does not necessarily make it determinately better or worse than the other. If a person faces a sequence of choices between incommensurate options, she may end up with a worse options than she could have had, even though none of her choices are irrational. Yet it seems that rationality should save her from this bad outcome. This is (...)
    Download  
     
    Export citation  
     
    Bookmark   24 citations  
  47. Democratic Values: A Better Foundation for Public Trust in Science.S. Andrew Schroeder - 2021 - British Journal for the Philosophy of Science 72 (2):545-562.
    There is a growing consensus among philosophers of science that core parts of the scientific process involve non-epistemic values. This undermines the traditional foundation for public trust in science. In this article I consider two proposals for justifying public trust in value-laden science. According to the first, scientists can promote trust by being transparent about their value choices. On the second, trust requires that the values of a scientist align with the values of an individual member of the (...)
    Download  
     
    Export citation  
     
    Bookmark   11 citations  
  48. Persons and the Satisfaction of Preferences: Problems in the Rational Kinematics of Values.Duncan MacIntosh - 1993 - Journal of Philosophy 90 (4):163-180.
    If one can get the targets of one's current wants only by acquiring new wants (as in the Prisoner's Dilemma), is it rational to do so? Arguably not. For this could justify adopting unsatisfiable wants, violating the rational duty to maximize one's utility. Further, why cause a want's target if one will not then want it? And people "are" their wants. So if these change, people will not survive to enjoy their wants' targets. I reply that one rationally need not (...)
    Download  
     
    Export citation  
     
    Bookmark   4 citations  
  49. The Problem of Mental Action.Thomas Metzinger - 2017 - Philosophy and Predicitive Processing.
    In mental action there is no motor output to be controlled and no sensory input vector that could be manipulated by bodily movement. It is therefore unclear whether this specific target phenomenon can be accommodated under the predictive processing framework at all, or if the concept of “active inference” can be adapted to this highly relevant explanatory domain. This contribution puts the phenomenon of mental action into explicit focus by introducing a set of novel conceptual instruments and developing a first (...)
    Download  
     
    Export citation  
     
    Bookmark   21 citations  
  50. Education, Values and Authority: A Semiotic View.Eetu Pikkarainen - 2014 - In Inna Semetsky & Andrew Stables (eds.), Pedagogy and Edusemiotics: theoretical challenges/practical opportunities. Sense Publisher. pp. 91-105.
    How can we theoretically and philosophically study the problem of values and authority in the context of education? The chapter uses the framework of action theoretical semiotics developed mainly on the conceptual structures of Greimassian semiotic theory. This detailed and elaborated theory of human discourse (utilized usually in terms of literary and “cultural” texts) will be expanded by biosemiotic and Peircean points of view to fit in the special problem area of education as transformation or extension from the (...)
    Download  
     
    Export citation  
     
    Bookmark   1 citation  
1 — 50 / 991