The no-free-lunch theorems promote a skeptical conclusion that all possible machine learning algorithms equally lack justification. But how could this leave room for a learning theory, that shows that some algorithms are better than others? Drawing parallels to the philosophy of induction, we point out that the no-free-lunch results presuppose a conception of learning algorithms as purely data-driven. On this conception, every algorithm must have an inherent inductive bias, that wants justification. We argue that many standard learning algorithms should rather (...) be understood as model-dependent: in each application they also require for input a model, representing a bias. Generic algorithms themselves, they can be given a model-relative justification. (shrink)

Scriven’s paradox of predictability arises from the combination of two ideas: first, that everything in a deterministic universe is, in principle, predictable; second, that it is possible to create a system that falsifies any prediction that is made of it. Recently, the paradox has been used by Rummens and Cuypers to argue that there is a fundamental difference between embedded and external predictors; and by Ismael to argue against a governing conception of laws. The present paper defends a new diagnosis (...) of the roots of the paradox. First, it is argued that the unpredictability has to be understood in the light of Turing’s famous results about computability, in particular his proof that there is no solution to the ‘halting problem.’ This allows us to see that previous analyses of the paradox were either mistaken or not fully adequate. Second, the sense of paradox that nevertheless remains is traced to the idea that rational behaviour is not dependent on contingent environmental circumstances: that it is always up to us to engage in activities such as rational prediction or rational belief. The paradox of predictability teaches us that this idea, natural though it may be, is mistaken. (shrink)

Wenmackers and Romeijn (2016) formalize ideas going back to Shimony (1970) and Putnam (1963) into an open-minded Bayesian inductive logic, that can dynamically incorporate statistical hypotheses proposed in the course of the learning process. In this paper, we show that Wenmackers and Romeijn’s proposal does not preserve the classical Bayesian consistency guarantee of merger with the true hypothesis. We diagnose the problem, and offer a forward-looking open-minded Bayesians that does preserve a version of this guarantee.

The global relation between logical empiricism and American pragmatism is one of the more difficult problems in history of philosophy. In this paper I’d like to take a local perspective and concentrate on the details that concern the vicissitudes of a philosopher who played an important role in the encounter of logical empiricism and American pragmatism, namely, Ernest Nagel. In this paper, I want to explore some aspects of Nagel’s changing attitude towards the then „new“ logical-empiricist philosophy. In the beginning (...) Nagel welcomed logical empiricism whole-heartedly. This early enthusiasm did not last. At the end of his philosophical career Nagel’s early positive attitude towards logical empiricism shown in the 1930s had been replaced by a much more reserved one. Nagel’s growing dissatisfaction with the Carnapian version of logical empiricist philosophy was clearly expressed in Nagel’s criticism of Carnap’s inductive logic and more generally in his last book Teleology Revisited and Other Essays on History and Philosophy of Science. There he critized harshly Carnap’s philosophy of science in general as ahistoric and non-pragmatist. One of the distinctive features of Nagel’s philosophy of science is the emphasis that he put on the role of history of science for philosophy of science. A compelling evidence for this attitude are his works on the history and philosophy of geometry and algebra One may say that Carnap and Nagel represented opposed possibilities of how the profession of a philosopher of science could be understood: Carnap as a „conceptual engineer“ was engaged in the task of inventing the conceptual tools for a better theoretical understanding of science, while Nagel was to be considered more as a „public intellectual“ engaged in the project of realizing a more rational and enlightened society. (shrink)