Synonyms
Definition
Occam’s Razor is the maxim that “entities are not to be multiplied beyond necessity,” or as it is often interpreted in the modern context “of two hypotheses H and H’, both of which explain E, the simpler is to be preferred” (Good 1977)
Motivation and Background
Most attempts to learn a model from data confront the problem that there will be many models that are consistent with the data. In order to learn a single model, a choice must be made between the available models. The factors taken into account by a learner in choosing between models are called its learning biases (Mitchell 1980). A preference for simple models is a common learning bias and is embodied in many learning techniques including pruning, minimum message length, and minimum description length. Regularization is also sometimes viewed as an application of Occam’s razor.
Occam’s razor is an imperative, rather than a proposition. That is, it is neither true nor false. Rather, it is a call...
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Recommended Reading
Blumer A, Ehrenfeucht A, Haussler D, Warmuth MK (1987) Occam’s razor. Inf Process Lett 24(6): 377–380
Domingos P (1999) The role of Occam’s razor in knowledge discovery. Data Min Knowl Discov 3(4):409–425
Good IJ (1977) Explicativity: a mathematical theory of explanation with statistical applications. Proc R Soc Lond Ser A 354:303–330
Mitchell TM (1980) The need for biases in learning generalizations. Department of computer science, Technical report CBM-TR-117, Rutgers University
Webb GI (1996) Further experimental evidence against the utility Of occams razor. J Artif Intell Res 4:397–417. AAAI Press, Menlo Park
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer Science+Business Media New York
About this entry
Cite this entry
Webb, G.I. (2017). Occam’s Razor. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning and Data Mining. Springer, Boston, MA. https://doi.org/10.1007/978-1-4899-7687-1_614
Download citation
DOI: https://doi.org/10.1007/978-1-4899-7687-1_614
Published:
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4899-7685-7
Online ISBN: 978-1-4899-7687-1
eBook Packages: Computer ScienceReference Module Computer Science and Engineering