Algorithmic information theory gives an idealized notion of compressibility that is often presented as an objective measure of simplicity. It is suggested at times that Solomonoff prediction, or algorithmic information theory in a predictive setting, can deliver an argument to justify Occam’s razor. This article explicates the relevant argument and, by converting it into a Bayesian framework, reveals why it has no such justificatory force. The supposed simplicity concept is better perceived as a specific inductive assumption, the assumption of (...) effectiveness. It is this assumption that is the characterizing element of Solomonoff prediction and wherein its philosophical interest lies. (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)

Algorithmic probability and friends: Proceedings of the Ray Solomonoff 85th memorial conference is a collection of original work and surveys. The Solomonoff 85th memorial conference was held at Monash University's Clayton campus in Melbourne, Australia as a tribute to pioneer, Ray Solomonoff, honouring his various pioneering works - most particularly, his revolutionary insight in the early 1960s that the universality of Universal Turing Machines could be used for universal Bayesian prediction and artificial intelligence. This work continues to (...) increasingly influence and under-pin statistics, econometrics, machine learning, data mining, inductive inference, search algorithms, data compression, theories of intelligence and philosophy of science - and applications of these areas. Ray not only envisioned this as the path to genuine artificial intelligence, but also, still in the 1960s, anticipated stages of progress in machine intelligence which would ultimately lead to machines surpassing human intelligence. Ray warned of the need to anticipate and discuss the potential consequences - and dangers - sooner rather than later. Possibly foremostly, Ray Solomonoff was a fine, happy, frugal and adventurous human being of gentle resolve who managed to fund himself while electing to conduct so much of his paradigm-changing research outside of the university system. The volume contains 35 papers pertaining to the abovementioned topics in tribute to Ray Solomonoff and his legacy. (shrink)

Sequential decision theory formally solves the problem of rational agents in uncertain worlds if the true environmental prior probability distribution is known. Solomonoff's theory of universal induction formally solves the problem of sequence prediction for unknown prior distribution. We combine both ideas and get a parameter-free theory of universal Artificial Intelligence. We give strong arguments that the resulting AIXI model is the most intelligent unbiased agent possible. We outline how the AIXI model can formally solve a number of (...) problem classes, including sequence prediction, strategic games, function minimization, reinforcement and supervised learning. The major drawback of the AIXI model is that it is uncomputable. To overcome this problem, we construct a modified algorithm AIXItl that is still effectively more intelligent than any other time t and length l bounded agent. The computation time of AIXItl is of the order t x 2^l. The discussion includes formal definitions of intelligence order relations, the horizon problem and relations of the AIXI theory to other AI approaches. (shrink)

An unknown process is generating a sequence of symbols, drawn from an alphabet, A. Given an initial segment of the sequence, how can one predict the next symbol? Ray Solomonoff’s theory of inductive reasoning rests on the idea that a useful estimate of a sequence’s true probability of being outputted by the unknown process is provided by its algorithmic probability (its probability of being outputted by a species of probabilistic Turing machine). However algorithmic probability is a “semimeasure”: i.e., the (...) sum, over all x∈A, of the conditional algorithmic probabilities of the next symbol being x, may be less than 1. Prevailing wisdom has it that algorithmic probability must be normalized, to eradicate this semimeasure property, before it can yield acceptable probability estimates. This paper argues, to the contrary, that the semimeasure property contributes substantially to the power and scope of an algorithmic-probability-based theory of induction, and that normalization is unnecessary. (shrink)

I will propose the notion that the universe is digital, not as a claim about what the universe is made of but rather about the way it unfolds. Central to the argument will be the concepts of symmetry breaking and algorithmic probability, which will be used as tools to compare the way patterns are distributed in our world to the way patterns are distributed in a simulated digital one. These concepts will provide a framework for a discussion of the informational (...) nature of reality. I will argue that if the universe were analog, then the world would likely be random, making it largely incomprehensible. The digital model has, however, an inherent beauty in its imposition of an upper limit and in the convergence in computational power to a maximal level of sophistication. Even if deterministic, that it is digital doesnít mean that the world is trivial or predictable, but rather that it is built up from operations that at the lowest scale are very simple but that at a higher scale look complex and even random, though only in appearance. (shrink)

I argue that Newtonian-style deduction-from-the-phenomena arguments should only carry conviction when they yield unexpectedly simple conclusions. That in that case they do establish higher rational probabilities for the theories they lead to than for any known or easily constructible rival theories. However I deny that such deductive justifications yield high absolute rational probabilities, and argue that the history of physics suggests that there are always other not-yet-known simpler theories with higher rational probabilities on all the original evidence, and that these (...) later turn out closer to the truth. My analyses rely on the modern Solomonoff-Levin solution to the problem of constructing a mathematically and philosophically acceptable inductive logic. (shrink)

Of all the sub-disciplines of philosophy, the philosophy of science has perhaps the most privileged relationship to information theory. This relationship has been forged through a common interest in themes like induction, probability, confirmation, simplicity, non-ad hocness, unification and, more generally, ontology. It also has historical roots. One of the founders of algorithmic information theory, Ray Solomonoff, produced his seminal work on inductive inference as a direct result of grappling with problems first encountered as a student of the (...) influential philosopher of science Rudolf Carnap. There are other such historical connections between the two fields. Alas, there is no space to explore them here. Instead this essay will restrict its attention to a broad and accessible overview of the aforementioned common themes, which, given their nature, mandate an emphasis on AIT as opposed to general information theory. (shrink)

In this thesis I investigate the theoretical possibility of a universal method of prediction. A prediction method is universal if it is always able to learn from data: if it is always able to extrapolate given data about past observations to maximally successful predictions about future observations. The context of this investigation is the broader philosophical question into the possibility of a formal specification of inductive or scientific reasoning, a question that also relates to modern-day speculation about a fully automatized (...) data-driven science. I investigate, in particular, a proposed definition of a universal prediction method that goes back to Solomonoff and Levin. This definition marks the birth of the theory of Kolmogorov complexity, and has a direct line to the information-theoretic approach in modern machine learning. Solomonoff's work was inspired by Carnap's program of inductive logic, and the more precise definition due to Levin can be seen as an explicit attempt to escape the diagonal argument that Putnam famously launched against the feasibility of Carnap's program. The Solomonoff-Levin definition essentially aims at a mixture of all possible prediction algorithms. An alternative interpretation is that the definition formalizes the idea that learning from data is equivalent to compressing data. In this guise, the definition is often presented as an implementation and even as a justification of Occam's razor, the principle that we should look for simple explanations. The conclusions of my investigation are negative. I show that the Solomonoff-Levin definition fails to unite two necessary conditions to count as a universal prediction method, as turns out be entailed by Putnam's original argument after all; and I argue that this indeed shows that no definition can. Moreover, I show that the suggested justification of Occam's razor does not work, and I argue that the relevant notion of simplicity as compressibility is already problematic itself. (shrink)

Putnam construed the aim of Carnap’s program of inductive logic as the specification of a “universal learning machine,” and presented a diagonal proof against the very possibility of such a thing. Yet the ideas of Solomonoff and Levin lead to a mathematical foundation of precisely those aspects of Carnap’s program that Putnam took issue with, and in particular, resurrect the notion of a universal mechanical rule for induction. In this paper, I take up the question whether the (...) class='Hi'>Solomonoff–Levin proposal is successful in this respect. I expose the general strategy to evade Putnam’s argument, leading to a broader discussion of the outer limits of mechanized induction. I argue that this strategy ultimately still succumbs to diagonalization, reinforcing Putnam’s impossibility claim. (shrink)

An a priori semimeasure (also known as “algorithmic probability” or “the Solomonoff prior” in the context of inductive inference) is defined as the transformation, by a given universal monotone Turing machine, of the uniform measure on the infinite strings. It is shown in this paper that the class of a priori semimeasures can equivalently be defined as the class of transformations, by all compatible universal monotone Turing machines, of any continuous computable measure in place of the uniform measure. Some (...) consideration is given to possible implications for the association of algorithmic probability with certain foundational principles of statistics. (shrink)

The framework of Solomonoff prediction assigns prior probability to hypotheses inversely proportional to their Kolmogorov complexity. There are two well-known problems. First, the Solomonoff prior is relative to a choice of Universal Turing machine. Second, the Solomonoff prior is not computable. However, there are responses to both problems. Different Solomonoff priors converge with more and more data. Further, there are computable approximations to the Solomonoff prior. I argue that there is a tension between these two (...) responses. This is because computable approximations to Solomonoff prediction do not always converge. (shrink)

Aristotle said that induction (epagōgē) is a proceeding from particulars to a universal, and the definition has been conventional ever since. But there is an ambiguity here. Induction in the Scholastic and the (so-called) Humean tradition has presumed that Aristotle meant going from particular statements to universal statements. But the alternate view, namely that Aristotle meant going from particular things to universal ideas, prevailed all through antiquity and then again from the time of Francis Bacon until the mid-nineteenth (...) century. Recent scholarship is so steeped in the first-mentioned tradition that we have virtually forgotten the other. In this essay McCaskey seeks to recover that alternate tradition, a tradition whose leading theoreticians were William Whewell, Francis Bacon, Socrates, and in fact Aristotle himself. The examination is both historical and philosophical. The first part of the essay fills out the history. The latter part examines the most mature of the philosophies in the Socratic tradition, specifically Bacon’s and Whewell’s. After tracing out this tradition, McCaskey shows how this alternate view of induction is indeed employed in science, as exemplified by several instances taken from actual scientific practice. In this manner, McCaskey proposes to us that the Humean problem of induction is merely an artifact of a bad conception of induction and that a return to the Socratic conception might be warranted. (shrink)

Itself a pleasure to read, this book is full of inventive arguments and conveys Millgram's bold thesis with elegance and force.

Although epistemic values have become widely accepted as part of scientific reasoning, non-epistemic values have been largely relegated to the "external" parts of science (the selection of hypotheses, restrictions on methodologies, and the use of scientific technologies). I argue that because of inductive risk, or the risk of error, non-epistemic values are required in science wherever non-epistemic consequences of error should be considered. I use examples from dioxin studies to illustrate how non-epistemic consequences of error can and should be considered (...) in the internal stages of science: choice of methodology, characterization of data, and interpretation of results. (shrink)

I present a new thermodynamic argument for the existence of God. Naturalistic physics provides evidence for the failure of induction, because it provides evidence that the past is not at all what you think it is, and your existence is just a momentary fluctuation. The fact that you are not a momentary fluctuation thus provides evidence for the existence of God – God would ensure that the past is roughly what we think it is, and you have been in (...) existence for roughly the amount of time you think you have. I don’t have a definitive way for the atheist to refute this argument, but I give one suggestion that relies on physics-based simplicity considerations. I close with an epistemological discussion of self-undermining arguments. (shrink)

Most social policies cannot be defended without making inductive inferences. For example, consider certain arguments for racial profiling and affirmative action, respectively. They begin with statistics about crime or socioeconomic indicators. Next, there is an inductive step in which the statistic is projected from the past to the future. Finally, there is a normative step in which a policy is proposed as a response in the service of some goal—for example, to reduce crime or to correct socioeconomic imbalances. In comparison (...) to the normative step, the inductive step of a policy defense may seem trivial. We argue that this is not so. Satisfying the demands of the inductive step is difficult, and doing so has important but underappreciated implications for the normative step. In this paper, we provide an account of induction in social contexts and explore its implications for policy. Our account helps to explain which normative principles we ought to accept, and as a result it can explain why it is acceptable to make inferences involving race in some contexts (e.g., in defense of affirmative action) but not in others (e.g., in defense of racial profiling). (shrink)

This paper formulates some paradoxes of inductive knowledge. Two responses in particular are explored: According to the first sort of theory, one is able to know in advance that certain observations will not be made unless a law exists. According to the other, this sort of knowledge is not available until after the observations have been made. Certain natural assumptions, such as the idea that the observations are just as informative as each other, the idea that they are independent, and (...) that they increase your knowledge monotonically (among others) are given precise formulations. Some surprising consequences of these assumptions are drawn, and their ramifications for the two theories examined. Finally, a simple model of inductive knowledge is offered, and independently derived from other principles concerning the interaction of knowledge and counterfactuals. (shrink)

The hard problem of induction is to argue without begging the question that inductive inference, applied properly in the proper circumstances, is conducive to truth. A recent theorem seems to show that the hard problem has a deductive solution. The theorem, provable in ZFC, states that a predictive function M exists with the following property: whatever world we live in, M ncorrectly predicts the world’s present state given its previous states at all times apart from a well-ordered subset. On (...) the usual model of time a well-ordered subset is small relative to the set of all times. M’s existence therefore seems to provide a solution to the hard problem. My paper argues for two conclusions. First, the theorem does not solve the hard problem of induction. More positively though, it solves a version of the problem in which the structure of time is given modulo our choice of set theory. (shrink)

Views which deny that there are necessary connections between distinct existences have often been criticized for leading to inductive skepticism. If there is no glue holding the world together then there seems to be no basis on which to infer from past to future. However, deniers of necessary connections have typically been unconcerned. After all, they say, everyone has a problem with induction. But, if we look at the connection between induction and explanation, we can develop the problem (...) of induction in a way that hits deniers of necessary connections, but not their opponents. The denier of necessary connections faces an `internal' problem with induction -- skepticism about important inductive inferences naturally flows from their position in a way that it doesn't for those who accept necessary connections. This is a major problem, perhaps a fatal one, for the denial of necessary connections. (shrink)

How induction was understood took a substantial turn during the Renaissance. At the beginning, induction was understood as it had been throughout the medieval period, as a kind of propositional inference that is stronger the more it approximates deduction. During the Renaissance, an older understanding, one prevalent in antiquity, was rediscovered and adopted. By this understanding, induction identifies defining characteristics using a process of comparing and contrasting. Important participants in the change were Jean Buridan, humanists such as (...) Lorenzo Valla and Rudolph Agricola, Paduan Aristotelians such as Agostino Nifo, Jacopo Zabarella, and members of the medical faculty, writers on philosophy of mind such as the Englishman John Case, writers of reasoning handbooks, and Francis Bacon. (shrink)

If the total state of the universe is encodable by a real number, Hardin and Taylor have proved that there is a solution to one version of the problem of induction, or at least a solution to a closely related epistemological problem. Is this philosophical application of the Hardin-Taylor result modest enough? The paper advances grounds for doubt. [A longer and more detailed sequel to this paper, 'Proving Induction', was published in the Australasian Journal of Logic in 2011.].

The standard backward-induction reasoning in a game like the centipede assumes that the players maintain a common belief in rationality throughout the game. But that is a dubious assumption. Suppose the first player X didn't terminate the game in the first round; what would the second player Y think then? Since the backwards-induction argument says X should terminate the game, and it is supposed to be a sound argument, Y might be entitled to doubt X's rationality. Alternatively, Y (...) might doubt that X believes Y is rational, or that X believes Y believes X is rational, or Y might have some higher-order doubt. X’s deviant first move might cause a breakdown in common belief in rationality, therefore. Once that goes, the entire argument fails. The argument also assumes that the players act rationally at each stage of the game, even if this stage could not be reached by rational play. But it is also dubious to assume that past irrationality never exerts a corrupting influence on present play. However, the backwards-induction argument can be reconstructed for the centipede game on a more secure basis.1 It may be implausible to assume a common belief in rationality throughout the game, however the game might go, but the argument requires less than this. The standard idealisations in game theory certainly allow us to assume a common belief in rationality at the beginning of the game. They also allow us to assume this common belief persists so long as no one makes an irrational move. That is enough for the argument to go through. (shrink)

Inductive methods can be used to estimate the accuracies of inductive methods. Call a method immodest if it estimates that it is at least as accurate as any of its rivals. It would be unreasonable to adopt any but an immodest method. Under certain assumptions, exactly one of Carnap's lambda-methods is immodest. This may seem to solve the problem of choosing among the lambda-methods; but sometimes the immodest lambda-method is λ =0, which it would not be reasonable to adopt. We (...) should therefore reconsider the assumptions that led to this conclusion: for instance, the measure of accuracy. (shrink)

From what norms does the ethics of belief derive its oughts, its attributions of virtues and vices, responsibilities and irresponsibilities, its permissioning and censuring? Since my inductive risk account is inspired by pragmatism, and this method understands epistemology as the theory of inquiry, the paper will try to explain what the aims and tasks are for an ethics of belief, or project of guidance, which best fits with this understanding of epistemology. More specifically, this chapter approaches the ethics of belief (...) from a focus on responsible risk management, where doxastic responsibility is understood in terms of the degree of riskiness of agents’ doxastic strategies, a riskiness which is in turn most objectively measured through accordance or violation of inductive norms. Doxastic responsibility is attributable to agents on the basis of how epistemically risky was the process or strategy of inquiry salient in the etiology of their belief or in their maintenance of what one believes or accepts. (shrink)

Category-based induction is an inferential mechanism that uses knowledge of conceptual relations in order to estimate how likely is for a property to be projected from one category to another. During the last decades, psychologists have identified several features of this mechanism, and they have proposed different formal models of it. In this article; we propose a new mathematical model for category-based induction based on distances on conceptual spaces. We show how this model can predict most of the (...) properties of this kind of reasoning while providing a solid theoretical foundation for it. We also show that it subsumes some of the previous models proposed in the literature and that it generates new predictions. (shrink)

An account of inductive inference is presented which addresses both its epistemological and metaphysical dimensions. It is argued that inductive knowledge is possible by virtue of the fit between our innate psychological capacities and the causal structure of the world.

As our epistemic ambitions grow, the common and scientific endeavours are becoming increasingly dependent on Machine Learning (ML). The field rests on a single experimental paradigm, which consists of splitting the available data into a training and testing set and using the latter to measure how well the trained ML model generalises to unseen samples. If the model reaches acceptable accuracy, an a posteriori contract comes into effect between humans and the model, supposedly allowing its deployment to target environments. Yet (...) the latter part of the contract depends on human inductive predictions or generalisations, which infer a uniformity between the trained ML model and the targets. The paper asks how we justify the contract between human and machine learning. It is argued that the justification becomes a pressing issue when we use ML to reach ‘elsewheres’ in space and time or deploy ML models in non-benign environments. The paper argues that the only viable version of the contract can be based on optimality (instead of on reliability which cannot be justified without circularity) and aligns this position with Schurz’s optimality justification. It is shown that when dealing with inaccessible/unstable ground-truths (‘elsewheres’ and non-benign targets), the optimality justification undergoes a slight change, which should reflect critically on our epistemic ambitions. Therefore, the study of ML robustness should involve not only heuristics that lead to acceptable accuracies on testing sets. The justification of human inductive predictions or generalisations about the uniformity between ML models and targets should be included as well. Without it, the assumptions about inductive risk minimisation in ML are not addressed in full. (shrink)

I review prominent historical arguments against scientific realism to indicate how they display a systematic overshooting in the conclusions drawn from the historical evidence. The root of the overshooting can be located in some critical, undue presuppositions regarding realism. I will highlight these presuppositions in connection with both Laudan’s ‘Old induction’ and Stanford’s New induction, and then delineate a minimal realist view that does without the problematic presuppositions.

In recent years, the argument from inductive risk against value free science has enjoyed a revival. This paper investigates and clarifies this argument through means of a case-study: neonicitinoid research. Sect. 1 argues that the argument from inductive risk is best conceptualised as a claim about scientists’ communicative obligations. Sect. 2 then shows why this argument is inapplicable to “public communication”. Sect. 3 outlines non-epistemic reasons why non-epistemic values should not play a role in public communicative contexts. Sect. 4 analyses (...) the implications of these arguments both for the specific case of neonicitinoid research and for understanding the limits of the argument from inductive risk. Sect. 5 sketches the broader implications of my claims for understanding the “Value Free Ideal” for science. (shrink)

Safety accounts of knowledge claim, roughly, that knowledge that p requires that one's belief that p could not have easily been false. Such accounts have been very popular in recent epistemology. However, one serious problem safety accounts have to confront is to explain why certain lottery‐related beliefs are not knowledge, without excluding obvious instances of inductive knowledge. We argue that the significance of this objection has hitherto been underappreciated by proponents of safety. We discuss Duncan Pritchard's recent solution to the (...) problem and argue that it fails. More importantly, the problem reaches deeper and poses a threat to any current safety accounts that require a belief's modal stability in close possibilities (as well as safety accounts that appeal to ‘normality’). We end by arguing that ways out of the problem require substantial reconstruction for a safety‐based account of knowledge. (shrink)

This article considers two arguments that purport to show that inductive reasoning is unjustified: the argument adduced by Sextus Empiricus and the (better known and more formidable) argument given by Hume in the Treatise. While Sextus’ argument can quite easily be rebutted, a close examination of the premises of Hume’s argument shows that they are seemingly cogent. Because the sceptical claim is very unintuitive, the sceptical argument constitutes a paradox. And since attributions of justification are theoretical, and the claim that (...) they are never (or seldom) true isn’t preposterous, the correct response to the paradox may well be to admit that the sceptic has exposed our error in making them. (shrink)

In ‘Induction and Natural Kinds’, I proposed a solution to the problem of induction according to which our use of inductive inference is reliable because it is grounded in the natural kind structure of the world. When we infer that unobserved members of a kind will have the same properties as observed members of the kind, we are right because all members of the kind possess the same essential properties. The claim that the existence of natural kinds is (...) what grounds reliable use of induction is based on an inference to the best explanation of the success of our inductive practices. As such, the argument for the existence of natural kinds employs a form of ampliative inference. But induction is likewise a form of ampliative inference. Given both of these facts, my account of the reliability of induction is subject to the objection that it provides a circular justification of induction, since it employs an ampliative inference to justify an ampliative inference. In this paper, I respond to the objection of circularity by arguing that what justifies induction is not the inference to the best explanation of its reliability. The ground of induction is the natural kinds themselves. (shrink)

Backward induction has been the standard method of solving finite extensive-form games with perfect information, notwithstanding the fact that this procedure leads to counter-intuitive results in various games. However, beginning in the late eighties, the method of backward induction became an object of criticism. It is claimed that the assumptions needed for its defence are quite implausible, if not incoherent. It is therefore natural to ask for the justification of backward induction: Can one show that rational players (...) who know the structure of the game, have trust in each other's practical rationality and reason correctly, will act in accordance in backward induction? Several researchers have tried a justification of this kind, but the argument presented in Robert Aumann's paper from 1995 is perhaps the most well-known and influential attempt to provide such a justification. Clausing provides a sustained discussion of the justification problem for backward induction. It is an excellent work and the criticism I will present below does not detract from this evaluation: the issues discussed by the author are complex and it is difficult to get everything right. Furthermore, I hope that the criticism to be presented may be instructive; Even though it has not been Clausing's intention, his logical reconstruction of Aumann's defence of backward induction allows us to see very clearly what is wrong with that argument. It also provides us with * This paper was presented at workshops in Lund and in Uppsala, in the Fall of 1999. I am indebted to the participants for their useful comments. The work on the paper was supported by a generous research grant from The Bank of Sweden Tercentenary Foundation. (shrink)

The argument from inductive risk is considered to be one of the strongest challenges for value-free science. A great part of its appeal lies in the idea that even an ideal epistemic agent—the “perfect scientist” or “scientist qua scientist”—cannot escape inductive risk. In this paper, I scrutinize this ambition by stipulating an idealized Bayesian decision setting. I argue that inductive risk does not show that the “perfect scientist” must, descriptively speaking, make non-epistemic value-judgements, at least not in a way that (...) undermines the value-free ideal. However, the argument is more successful in showing that there are cases where the “perfect scientist” should, normatively speaking, use non-epistemic values. I also show that this is possible without creating problems of illegitimate prescription and wishful thinking. Thus, while inductive risk does not refute value-freedom completely, it still represents a powerful critique of value-free science. (shrink)

In this article, I argue that arguments from the history of science against scientific realism, like the arguments advanced by P. Kyle Stanford and Peter Vickers, are fallacious. The so-called Old Induction, like Vickers's, and New Induction, like Stanford's, are both guilty of confirmation bias—specifically, of cherry-picking evidence that allegedly challenges scientific realism while ignoring evidence to the contrary. I also show that the historical episodes that Stanford adduces in support of his New Induction are indeterminate between (...) a pessimistic and an optimistic interpretation. For these reasons, these arguments are fallacious, and thus do not pose a serious challenge to scientific realism. (shrink)

. Recent philosophers of science have not only revived the classical argument from inductive risk but extended it. I argue that some of the purported extensions do not fit cleanly within the schema of the original argument, and I discuss the problem of overdiagnosis of disease due to expanded disease definitions in order to show that there are some risks in the research process that are important and that very clearly fall outside of the domain of inductive risk. Finally, I (...) introduce the notion of epistemic risk in order to characterize such risks. (shrink)

The argument from inductive risk has been embraced by many as a successful account of the role of values in science that challenges the value-free ideal. We argue that it is not obvious that the argument from inductive risk actually undermines the value-free ideal. This is because the inductive risk argument endorses an assumption held by proponents of the value-free ideal: that contextual values never play an appropriate role in determining evidence. We show that challenging the value-free ideal ultimately requires (...) rejecting this assumption. (shrink)

The pessimistic induction plays an important role in the contemporary realism/anti-realism debate in philosophy of science. But there is some disagreement about the structure and aim of the argument. And a number of scholars have noted that there is more than one type of PI in the philosophical literature. I review four different versions of the PI. I aim to show that PIs have been appealed to by philosophers of science for a variety of reasons. Even some realists have (...) appealed to a PI. My goal is to advance our understanding of what the various PIs can teach us about science and the threat posed by PIs to scientific realism. (shrink)

Pure Inductive Logic is the study of rational probability treated as a branch of mathematical logic. This monograph, the first devoted to this approach, brings together the key results from the past seventy years, plus the main contributions of the authors and their collaborators over the last decade, to present a comprehensive account of the discipline within a single unified context.