Department of Philosophy

Permanent URI for this communityhttps://hdl.handle.net/11693/115654

Browse

Browsing Department of Philosophy by Subject "Akaike information criterion"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Akaike and the No Miracle Argument for scientific realism
    (Cambridge University Press, 2023-08-03) Fatollahi, Alireza; McCullagh, M.
    The “No Miracle Argument” for scientific realism contends that the only plausible explanation for the predictive success of scientific theories is their truthlikeness, but doesn’t specify what ‘truthlikeness’ means. I argue that if we understand ‘truthlikeness’ in terms of Kullback-Leibler (KL) divergence, the resulting realist thesis (RKL) is a plausible explanation for science’s success. Still, RKL probably falls short of the realist’s ideal. I argue, however, that the strongest version of realism that the argument can plausibly establish is RKL. The realist needs another argument for establishing a stronger realist thesis.
  • Thumbnail Image
    ItemOpen Access
    Predictivism and model selection
    (Springer Science and Business Media B.V., 2023-02-21) Fatollahi, Alireza
    There has been a lively debate in the philosophy of science over predictivism: the thesis that successfully predicting a given body of data provides stronger evidence for a theory than merely accommodating the same body of data. I argue for a very strong version of the thesis using statistical results on the so-called “model selection” problem. This is the problem of finding the optimal model (family of hypotheses) given a body of data. The key idea that I will borrow from the statistical literature is that the level of support a hypothesis, H, receives from a body of data, D, is inversely related to the number of adjustable parameters of the model from which H was constructed. I will argue that when D is not essential to the design of H (i.e., when it is predicted), the model to which H belongs has fewer adjustable parameters than when D is essential to the design of H (when it is accommodated). This, I argue, provides us with an argument for a very strong version of predictivism.