Can humans get arbitrarily capable reinforcement learning agents to do their bidding? Or will sufficiently capable RL agents always find ways to bypass their intended objectives by shortcutting their reward signal? This question impacts how far RL can be scaled, and whether alternative paradigms must be developed in order to build safe artificial general intelligence. In this paper, we study when an RL agent has an instrumental goal to tamper with its reward process, and describe design principles that prevent instrumental (...) goals for two different types of reward tampering. Combined, the design principles can prevent reward tampering from being an instrumental goal. The analysis benefits from causal influence diagrams to provide intuitive yet precise formalizations. (shrink)

A fundamental problem in artificial intelligence is that nobody really knows what intelligence is. The problem is especially acute when we need to consider artificial systems which are significantly different to humans. In this paper we approach this problem in the following way: we take a number of well known informal definitions of human intelligence that have been given by experts, and extract their essential features. These are then mathematically formalised to produce a general measure of intelligence for arbitrary machines. (...) We believe that this equation formally captures the concept of machine intelligence in the broadest reasonable sense. We then show how this formal definition is related to the theory of universal optimal learning agents. Finally, we survey the many other tests and definitions of intelligence that have been proposed for machines. (shrink)

ABSTRACTResearch that dissociates different types of processes within a given task using a processing tree approach suggests that attitudes may be acquired through evaluative conditioning in the absence of explicit encoding of CS-US pairings in memory. This research distinguishes explicit memory for the CS-US pairings from CS-liking acquired without encoding of CS-US pairs in explicit memory. It has been suggested that the latter effect may be due to an implicit misattribution process that is assumed to operate when US evocativeness is (...) low. In the present research, the latter assumption was supported neither by two high-powered experiments nor by complementary meta-analytic evidence, whereas evocativeness exerted an influence on explicit memory. This pattern of findings is inconsistent with the view that CS-liking acquired without encoding of CS-US pairs in explicit memory reflects an implicit misattribution process at learning. Hence, the underlying learning process is awaiting further empirical scrutiny. (shrink)

Das Verhaltnis von Spekulation und Vorstellung in Hegels enzyklopadischem System ist von einer eigentumlichen Spannung gekennzeichnet. Spekulation ist nicht nur die kritische Reinigung der Inhalte der Vorstellung von ihren Vorurteilen und Irrtumern, sondern zugleich Anerkennung des endlichen, vorstellenden Denkens innerhalb seiner Grenzen, sowohl auf dem Gebiet der Erkenntnistheorie als auch der Religion. Wie verhalten sich aber Spekulation und Vorstellung zueinander in verschiedenen Teilen des enzyklopadischen Systems, in der Logik und der Philosophie des Geistes? Diesem Zusammenhang gehen die Beitrage des Tagungsbandes (...) im Detail nach und uberprufen kritisch das enzyklopadische Zusammenspiel von Spekulation und Vorstellung in den Spharen des subjektiven und des absoluten Geistes. Mit Beitragen von: Kazimir Drilo, Markus Gabriel, Christoph Halbig, Gunnar Hindrichs, Axel Hutter, Franz Knappik, Anton Friedrich Koch, Christian Georg Martin, Jan Rohls, Rainer Schafer. (shrink)

Can an agent's intelligence level be negative? We extend the Legg-Hutter agent-environment framework to include punishments and argue for an affirmative answer to that question. We show that if the background encodings and Universal Turing Machine (UTM) admit certain Kolmogorov complexity symmetries, then the resulting Legg-Hutter intelligence measure is symmetric about the origin. In particular, this implies reward-ignoring agents have Legg-Hutter intelligence 0 according to such UTMs.

The Conference on Artificial General Intelligence is the only major conference series devoted wholly and specifically to the creation of AI systems possessing general intelligence at the human level and ultimately beyond. Its second installation, AGI-09, in Arlington, Virginia, March 6-9, 2009, attracted 67 paper submissions, which is a substantial increase from the previous year. Of these submissions, 33 (i.e., 49%) were accepted as full papers for presentation at the conference. Additional 13 papers were included as position papers. The program (...) also included a keynote address by Juergen Schmidhuber on 'The New AI', a post-conference workshop on 'The Future of AI', and a number of pre-conference tutorials on various topics related to AGI. (shrink)

The call for business practices that create benefits for companies, customers, and society is getting louder. This article analyzes a new implementation of such a win–win–win approach: the carrotmob. Activists and managers jointly organize a shopping flashmob in which consumers collectively purchase the products of a target company to reward its intent to act more socially responsible. Given that carrotmobs are only efficient if they are supported by a critical mass of consumers, a survey study of 337 young consumers explores (...) the critical drivers of carrotmob participation. Accordingly, object-oriented, personal, and social motives jointly determine carrotmob participation with social motives having the strongest impact. (shrink)

ABSTRACTRecent research into evaluative conditioning shows that information about the relationship between the conditioned and unconditioned stimuli can exert strong effects on the size and di...

Name der Zeitschrift: Zeitschrift für Religionswissenschaft Jahrgang: 26 Heft: 1 Seiten: 202-205.

Name der Zeitschrift: Zeitschrift für Religionswissenschaft Jahrgang: 26 Heft: 1 Seiten: 205-207.

Automated reasoning about uncertain knowledge has many applications. One difficulty when developing such systems is the lack of a completely satisfactory integration of logic and probability. We address this problem directly. Expressive languages like higher-order logic are ideally suited for representing and reasoning about structured knowledge. Uncertain knowledge can be modeled by using graded probabilities rather than binary truth-values. The main technical problem studied in this paper is the following: Given a set of sentences, each having some probability of being (...) true, what probability should be ascribed to other (query) sentences? A natural wish-list, among others, is that the probability distribution (i) is consistent with the knowledge base, (ii) allows for a consistent inference procedure and in particular (iii) reduces to deductive logic in the limit of probabilities being 0 and 1, (iv) allows (Bayesian) inductive reasoning and (v) learning in the limit and in particular (vi) allows confirmation of universally quantified hypotheses/sentences. We translate this wish-list into technical requirements for a prior probability and show that probabilities satisfying all our criteria exist. We also give explicit constructions and several general characterizations of probabilities that satisfy some or all of the criteria and various (counter) examples. We also derive necessary and sufficient conditions for extending beliefs about finitely many sentences to suitable probabilities over all sentences, and in particular least dogmatic or least biased ones. We conclude with a brief outlook on how the developed theory might be used and approximated in autonomous reasoning agents. Our theory is a step towards a globally consistent and empirically satisfactory unification of probability and logic. (shrink)

Understanding inductive reasoning is a problem that has engaged mankind for thousands of years. This problem is relevant to a wide range of fields and is integral to the philosophy of science. It has been tackled by many great minds ranging from philosophers to scientists to mathematicians, and more recently computer scientists. In this article we argue the case for Solomonoff Induction, a formal inductive framework which combines algorithmic information theory with the Bayesian framework. Although it achieves excellent theoretical results (...) and is based on solid philosophical foundations, the requisite technical knowledge necessary for understanding this framework has caused it to remain largely unknown and unappreciated in the wider scientific community. The main contribution of this article is to convey Solomonoff induction and its related concepts in a generally accessible form with the aim of bridging this current technical gap. In the process we examine the major historical contributions that have led to the formulation of Solomonoff Induction as well as criticisms of Solomonoff and induction in general. In particular we examine how Solomonoff induction addresses many issues that have plagued other inductive systems, such as the black ravens paradox and the confirmation problem, and compare this approach with other recent approaches. (shrink)

In lieu of an abstract, here is a brief excerpt of the content:Reviewed by:An Evangelical Adrift: The Making of John Henry Newman's Theology by Geertjan ZuijdwegtReinhard HütterAn Evangelical Adrift: The Making of John Henry Newman's Theology by Geertjan Zuijdwegt (Washington, DC: The Catholic University of America Press, 2022), xii + 365pp.What do we know of St. John Henry Newman's life, thought, and theological development before 1833, the year in which the Oxford Movement or Tractarian Movement was launched by Newman, John (...) Keble, and Richard Hurrell Froude? Newman, in his rightly famous and indispensable Apologia Pro Vita Sua, covers this period in one brief chapter offering at best a rough outline and some central signposts. A more nuanced and somewhat more expansive picture is emerging in Newman's Autobiographical Writings, published posthumously in 1956. Wilfrid Ward, in the two immense volumes of his 1912 Newman biography, dedicates one single chapter entitled "Life in the Church of England (1801–1845)" to the first forty-four years of Newman's life, and of those fifty-four pages, only thirty cover the time from Newman's birth in 1801 to the signal year 1833. Ian Ker, in his 1988 biography, arguably still the normative and unsurpassed scholarly Newman biography, acknowledged as such by admirers and critics alike, dedicates one chapter of fifty-three pages to the same time period. Finally, Geertjan Zuijdwegt, in his very well written and equally well-researched and well-argued monograph of 344 pages, offers a detailed and nuanced account of Newman's theological development from his teenage conversion to Calvinist Evangelicalism up to and including the beginnings of the Tractarian Movement from 1833 to 1835. One must go back to Maisie Ward's lovely portrait, Young Mr. Newman, from 1948, to find a book that covers roughly the same territory. Ward's captivating biographical portrait and Zuijdwegt's meticulously researched theological biography may be read with great benefit as mutually enlightening and complementary accounts—at least in the eyes of this reader. The latter's new work is, however, a commendable and indeed indispensable achievement of its own. For one, what lies between Ward's portrait of the young Newman and Zuijdwegt's reconstruction of Newman's [End Page 1097] early intellectual and theological development is Frank M. Turner's 2002 John Henry Newman: The Challenge to Evangelical Religion, a 750-page ad hominem attack on Newman's person and character, not only making Evangelicalism (and not theological and political liberalism) the alleged primary if not sole object of the post-Evangelical Newman's Tractarian and post-Tractarian, Catholic polemic, but also insinuating alongside in so many ways that Newman was dishonest about his work and thought, that he cared deeply for money or the security of a "living," that he was a closeted and repressed homosexual, and that he was lying about his conversion experiences—charges advanced by Charles Kingsley and other contemporaries of Newman's day and age and debunked—now warmed up und recycled under the guise of what pretended to be a massive scholarly study of intellectual history published by an Ivy League university press. Much ink has been spilled by Newman scholars in the years since Turner's book appeared to set the record straight, while others thought the time was ripe to form a "Turner school" of critical, iconoclastic Newman scholarship. Unsurprisingly, since 2002, there has been no dull moment in the small world of Newman scholarship, now divided between two well-defined camps, with "iconographers" (so called by the iconoclasts) on one side, and on the other those who aim at advancing a hard-nosed form of critical, detached historical Newman scholarship that at points has a hard time identifying the fine line between historical criticism and barely camouflaged detraction.Zuijdwegt's book is a considerable achievement first and foremost simply because he escapes to a commendably high degree, albeit not completely, the trench warfare between the warring camps of Newman scholarship. His unavoidable engagement with Turner's work is judicious, balanced, and overall, in a measured way, critical. He is far more frequently disagreeing with Turner's interpretation than agreeing with a limited number of interpretive points about Newman's early... (shrink)

Much of the research that has been carried out on occupational health and safety involves studies of regulatory practices. OHS studies linearly maintain that early legislations were of minimal consequence. Implementation is a two-tier structure—policy-making and enforcement. This article considers the main themes and findings of this body of research. It is structured around a “natural history” approach to understanding law. This approach regards law as a process which starts with the recognition of a problem demanding legal intervention and the (...) subsequent enactment of legislation. Several jurisdictions additionally encourage workers to join the enforcement procedure. Law being intrinsic to OHS, enforcement emerges as the key to explain its impact, while evidence suggests the role of sanctions on it. The article highlights areas that remain relatively uncharted and those that warrant more research. (shrink)

The article develops the thesis that spiritual objectivity constitutes an independent class of phenomena besides the physical and the mental. The concept of spiritual objectivity presents a solution for the mediation between the bodily and the conscious by further developing the insight of critical monism that individual action can neither be subsumed under the phenomena of the bodily outer world nor under the phenomena of the mental inner world. Referring to Gottlob Frege's thesis that what distinguishes a thought from other (...) phenomena is its capability of being true, the article develops the argument that the thought in its spiritual objectivity should be distinguished from the psychic inner world of ideas as well as from the physical exterior world of material objects. The possibility of rational recognition is dependent upon this genuinely spiritual mediation of inner and outer world, and its objective truth must be distinguished from the constitutive subjectivity of psychic phenomena. This specified concept of the spiritual is not intended to enlarge the substance dualism between the bodily and the mental once again but makes it possible to determine the in itself differentiated unity of the bodily and the mental, namely as spiritual unity of body and mind. (shrink)

The article develops the thesis that spiritual objectivity constitutes an independent class of phenomena besides the physical and the mental. The concept of spiritual objectivity presents a solution for the mediation between the bodily and the conscious by further developing the insight of critical monism that individual action can neither be subsumed under the phenomena of the bodily outer world nor under the phenomena of the mental inner world. Referring to Gottlob Frege's thesis that what distinguishes a thought from other (...) phenomena is its capability of being true, the article develops the argument that the thought in its spiritual objectivity should be distinguished from the psychic inner world of ideas as well as from the physical exterior world of material objects. The possibility of rational recognition is dependent upon this genuinely spiritual mediation of inner and outer world, and its objective truth must be distinguished from the constitutive subjectivity of psychic phenomena. This specified concept of the spiritual is not intended to enlarge the substance dualism between the bodily and the mental once again but makes it possible to determine the in itself differentiated unity of the bodily and the mental, namely as spiritual unity of body and mind. (shrink)

Shaping the Future maps out the ascetic practices of a Neitzschean way of life. Hutter argues that Nietzsche's doctrines are attempts and "temptations" that aim to provoke his free-spirited readers into changing themselves by putting philosophy into practice in their lives.