|Abstract||A ﬁnite data set is consistent with inﬁnitely many alternative theories. Scientiﬁc realists recommend that we prefer the simplest one. Anti-realists ask how a ﬁxed simplicity bias could track the truth when the truth might be complex. It is no solution to impose a prior probability distribution biased toward simplicity, for such a distribution merely embodies the bias at issue without explaining its eﬃcacy. In this note, I argue, on the basis of computational learning theory, that a ﬁxed simplicity bias is necessary if inquiry is to converge to the right answer eﬃciently, whatever the right answer might be. Eﬃciency is understood in the sense of minimizing the least ﬁxed bound on retractions or errors prior to convergence.|
|Keywords||No keywords specified (fix it)|
|Categories||No categories specified (fix it)|
|Through your library||Only published papers are available at libraries|
Similar books and articles
R. Sorensen (2011). Simpler Without a Simplest: Ockham's Razor Implies Epistemic Dilemmas. Analysis 71 (2):260-264.
Daniel Steel (2009). Testability and Ockham's Razor: How Formal and Statistical Learning Theory Converge in the New Riddle of Induction. Journal of Philosophical Logic 38 (5):471 - 489.
Arnold Zellner, Hugo A. Keuzenkamp & Michael McAleer (eds.) (2001). Simplicity, Inference and Modeling: Keeping It Sophisticatedly Simple. Cambridge University Press.
Kevin T. Kelly (2004). Justification as Truth-Finding Efficiency: How Ockham's Razor Works. Minds and Machines 14 (4):485-505.
Kevin Kelly (2004). Justification as Truth-Finding Efficiency: How Ockham's Razor Works. Minds and Machines 14 (4):485-505.
Added to index2009-01-28
Total downloads12 ( #93,337 of 549,065 )
Recent downloads (6 months)1 ( #63,185 of 549,065 )
How can I increase my downloads?