Online Learning of k-CNF Boolean Functions
نویسندگان
چکیده
This paper revisits the problem of learning a k-CNF Boolean function from examples, for fixed k, in the context of online learning under the logarithmic loss. We give a Bayesian interpretation to one of Valiant’s classic PAC learning algorithms, which we then build upon to derive three efficient, online, probabilistic, supervised learning algorithms for predicting the output of an unknown k-CNF Boolean function. We analyze the loss of our methods, and show that the cumulative log-loss can be upper bounded by a polynomial function of the size of each example.
منابع مشابه
Learning Boolean Functions in an In nite Attribute Space
This paper presents a theoretical model for learning Boolean functions in domains having a large, potentially innnite number of attributes. The model allows an algorithm to employ a rich vocabulary to describe the objects it encounters in the world without necessarily incurring time and space penalties so long as each individual object is relatively simple. We show that many of the basic Boolea...
متن کاملAnalysis of Case-Based Representability of Boolean Functions by Monotone Theory
Classi cation is one of major tasks in case-based reasoning(CBR) and many studies have been done for analyzing properties of case-based classi cation [1, 14, 10, 15, 12, 9, 13, 7]. However, these studies only consider numerical similarity measures whereas there are other kinds of similarity measure for di erent tasks. Among these measures, HYPO system [2, 3] in a legal domain uses a similarity ...
متن کاملAn incremental learning algorithm for constructing Boolean functions from positive and negative examples
This paper introduces an incremental algorithm for learning a Boolean function from examples. The functions are constructed in the disjunctive normal form (DNF) or the conjunctive normal form (CNF) and emphasis is placed in inferring functions with as few clauses as possible. This incremental algorithm can be combined with any existing algorithm that infers a Boolean function from examples. In ...
متن کاملEecient Read-restricted Monotone Cnf/dnf Dualization by Learning with Membership Queries Produced as Part of the Esprit Working Group in Neural and Computational Learning Ii, Neurocolt2 27150
We consider exact learning monotone CNF formulas in which each variable appears at most some constant k times (\read-k" monotone CNF). Let f : f0;1g n ! f0;1g be expressible as a read-k monotone CNF formula for some natural number k. We give an incremental output polynomial time algorithm for exact learning both the read-k CNF and (not necessarily read restricted) DNF descriptions of f. The alg...
متن کاملAlmost all monotone Boolean functions are polynomially learnable using membership queries
We consider exact learning or identification of monotone Boolean functions by only using membership queries. It is shown that almost all monotone Boolean functions are polynomially identifiable in the input number of variables as well as the output being the sum of the sizes of the CNF and DNF representations. 2001 Elsevier Science B.V. All rights reserved.
متن کامل