Filters








6 Hits in 4.7 sec

On the necessity of Occam algorithms

R. Board, L. Pitt
1990 Proceedings of the twenty-second annual ACM symposium on Theory of computing - STOC '90  
A number of restricted classes of representations are polynomially size-bounded, including k-decision-lists, k-term DNF formulas, k-clause CNF formulas, k-DNF formulas, and k-CNF formulas, where k is any  ...  There exist some classes, such as the class of formulas in disjunctive normal form (DNF) with at most k terms, that are not properly PAC-learnable (unless RP = NP), but are learnable in terms of another  ... 
doi:10.1145/100216.100223 dblp:conf/stoc/BoardP90 fatcat:xpvz3utodngorotuk5g2o2lsnq

On the necessity of Occam algorithms

Raymond Board, Leonard Pitt
1992 Theoretical Computer Science  
A number of restricted classes of representations are polynomially size-bounded, including k-decision-lists, k-term DNF formulas, k-clause CNF formulas, k-DNF formulas, and k-CNF formulas, where k is any  ...  There exist some classes, such as the class of formulas in disjunctive normal form (DNF) with at most k terms, that are not properly PAC-learnable (unless RP = NP), but are learnable in terms of another  ... 
doi:10.1016/0304-3975(92)90367-o fatcat:viby4shwenen5exxkm2yndxjqa

Min-max classifiers: Learnability, design and application

Ping-Fai Yang, Petros Maragos
1995 Pattern Recognition  
Several subclasses of thresholded min-max functions are shown to be learnable, generalizing the learnability results for the corresponding classes of Boolean functions.  ...  We studied supervised learning of these classifiers under the Probably Approximately Correct (PAC) model proposed by Valiant.  ...  Exchanging the roles of AND and OR in a DNF Boolean function transforms the latter into a conjunc-* In this paper, Boolean AND is denoted by the product symbol '.  ... 
doi:10.1016/0031-3203(94)00161-e fatcat:ewzeweoxfbfi3ek5535kaq3un4

Data Enrichment [chapter]

2017 Encyclopedia of Machine Learning and Data Mining  
Data preparation is the process of manipulating and organizing data prior to analysis.  ...  For discrete attributes, this typically means that the tree consists only of a single interior node (i.e., the root has only leaves as successor nodes).  ...  on the population, and is known under the name of Ewens sampling formula (Ewens 1972) .  ... 
doi:10.1007/978-1-4899-7687-1_979 fatcat:v22x3o5iuzaa5ghwx5fvqztpg4

Learning and Inductive Inference [chapter]

1982 The Handbook of Artificial Intelligence  
One important set of heuristics is based on the idea of the distribution of substrings in the language.  ...  Prior distributions.  ...  Examine the following context-free grammar, C, with This grammar, G, describes a language of simple sentences such as The boy held the puppy and The girl on the hill held a kitten.  ... 
doi:10.1016/b978-0-86576-091-2.50009-5 fatcat:whndkx7nhrdrrn3k7y2f7b3vti

Effective techniques for handling incomplete data using decision trees

Bhekisipho E.T.H. Twala
2005
The first contribution is the extensive simulations which study the impact of missing data on predictive accuracy of existing DTs which can cope with missing values, when missing values are in both the  ...  Decision Trees (DTs) have been recognized as one of the most successful formalisms for knowledge representation and reasoning and are currently applied to a variety of data mining or knowledge discovery  ...  This model is analogous to the LR model, except that the probability distribution of the response is multinomial instead of binomial and there are k equations instead of one.  ... 
doi:10.21954/ou.ro.0000e969 fatcat:pshr7eoawvbbzhspngggrk2apm