• There will be no vacation during the one-week camp program.
  **SS: Sahotra Sarkar; ME: Mehmet Elgin; ED: Erhan Demircioğlu; BF: Branden Fitelson; TB: Tolga Birdal

The Vienna Circle and the Confirmation of Scientific Theories (Sahotra Sarkar, University of Texas at Austin)
In the 1920s and 1930s the logical positivism of the Vienna Circle developed the major themes that dominated philosophy of science in the twentieth century. This set of lectures explores the evolution of Vienna Circle doctrine on confirmation from the 1920s through the 1950s, beginning with the use of verifiability as a criterion of both meaning and verification and ending with attempts to formulate an inductive logic in the 1950s. This evolution marked a transition from viewing the confirmation of scientific theories as a pragmatic task to viewing it as a formal project in epistemology. The major figures treated in historical order will include Schlick, Neurath, Carnap, Reichenbach, and Hempel, as well as their students.

Lectures on Popper, Kuhn and Lakatos (Mehmet Elgin, Muğla Sıtkı Koçman University, Muğla)
It is no exaggeration to say that Karl S. Popper and Thomas S. Kuhn are the most influential philosophers of science in the 20^th century. Imre Lakatos attempted to offer a theory of scientific method by synthesizing Popper and Kuhn’s views and there is somewhat informal agreement among philosophers of science that Lakatos’ theory of science is superior to those of Popper and Kuhn. In the lectures, we will focus on three books: Logic of Scientific Discovery by Popper, The Structure of Scientific Revolutions by Kuhn and Methodology of Scientific Research Programs by Lakatos. Popper’s book was a reaction to the views of logical empiricist that the first set of lectures will cover. Kuhn’s book was a reaction to both Popper and logical empiricists and Lakatos’ book was a reaction to Kuhn and Popper. In the lecture about Popper, we will focus on subtle details of Popper’s view that have often been misunderstood. In the lecture on Kuhn, we will focus on the incommensurability of rival scientific theories, rationality of theory-choice and Kuhn’s criticism of the realist interpretation of science. In the lecture on Lakatos, we will address the question of whether Lakatos’ theory is superior to that of Kuhn in giving a rational account of science.

Problem of Induction (Erhan Demircioğlu, Koç University, Istanbul)
The problem of induction is the problem of understanding the epistemic/evidential contribution our observations make to the beliefs that we form about the things that we have not yet observed. On the one hand, it is highly plausible that our observations have the potential to justify our beliefs about unobserved matters of facts. On the other hand, it is highly difficult to figure out the mechanism by which that sort of justification might work. A series of lectures in these meetings will directly address this famous problem in philosophy of science and investigate a number of attempts to solve it.

Probability, Induction, and Human Reasoning (Branden Fitelson, Northeastern University, Boston)
In these lectures, I will begin by introducing the basic logical and probabilistic concepts necessary for understanding contemporary inductive and statistical inference.  Then, I will discuss how these concepts can be used measure the strength of inductive (viz., non-deductive) arguments.  This will, in turn, lead to several applications in the philosophy of science and the psychology of human reasoning.  First, I will focus on some foundational paradoxes of inductive reasoning/confirmation (some “paradoxes of confirmation”).  And, then, I will discuss several so-called “reasoning fallacies” that have appeared in the literature on the psychology of human reasoning.  In both cases, our general probabilistic framework for assessing the strength of arguments will provide valuable philosophical and psychological insights.

Artificial Intelligence (Tolga Birdal, Imperial College London, London)
I will venture into the intricate philosophical underpinnings of inference within the realm of artificial intelligence (AI). In particular I will investigate the foundational concepts of point estimates and Bayesian Inference, illuminating how these approaches facilitate understanding and predictions about the world through AI models. Central to this exploration is the examination of posterior estimation methods, specifically variational inference and Markov Chain Monte Carlo (MCMC), which serve as pivotal tools in the AI toolkit for grappling with the complexities of probabilistic reasoning. Arguing that no single solution exists for the prediction problems, a critical aspect of our discourse revolves around the concepts of uncertainty and ambiguity in AI. These elements are inherent in real-world data and decision-making processes. The final segment focuses on how AI systems infer rules and patterns from data, shedding light on the philosophical questions surrounding the nature of knowledge, learning, and intelligence. By the end of this course, participants will gain a profound understanding of the philosophical dimensions of inference in AI, equipped with certain theoretical knowledge to navigate the complex and ever-evolving landscape of artificial intelligence.