Extending Occam's Razor

نویسندگان

  • Kevin S. Van Horn
  • Tony R. Martinez
چکیده

Occam's Razor states that, all other things being equal, the simpler of two possible hypotheses is to be preferred. A quanti ed version of Occam's Razor has been proven for the PAC model of learning, giving sample-complexity bounds for learning using what Blumer et al. call an Occam algorithm [1]. We prove an analog of this result for Haussler's more general learning model, which encompasses learning in stochastic situations, learning real-valued functions, etc.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Conditions for Occam's Razor Applicability and Noise Elimination

The Occam's razor principle suggests that among all the correct hypotheses, the simplest hypothesis is the one which best captures the structure of the problem domain and has the highest prediction accuracy when classifying new instances. This principle is implicitly used also for dealing with noise, in order to avoid overrtting a noisy training set by rule truncation or by pruning of decision ...

متن کامل

Occam's Razor & The Nature of Scientific Theories

he early 14th century philosopher William of Occam (Sahakian 1968, p. 116) is credited by historians for proposing that, in interpreting scientific data, the simplest explanation consistent with all the known facts is preferred as a working hypothesis over more complicated ones. In other words, scientific explanations should be pared down (as if using a knife to cut away unnecessary "fat" from ...

متن کامل

God Doesn't Always Shave with Occam's Razor - Learning When and How to Prune

The work shows how a meta-learning technique can be successfully applied to decide when to prune, how much pruning is appropriate and what the best pruning technique is for a given learning task.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1994