M. Krause. On the computational power of Boolean decision lists. In 19th Annual Symposium on Theoretical Aspects of Computer Science, pages 372--383, 2002.  Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Analysis of Data with Threshold Decision Lists - Martin Anthony December (Correct)
....set of finite sequences f = f 1 ; c 1 ) f 2 ; c 2 ) f r ; c r ) such that f i 2 K and c i 2 f0; 1g for 1 i r. The values of f are defined by f(y) c j where j = minfi j f i (y) 1g, or 0 if there are no j such that f j (y) 1. We call each f j a test (or, following Krause [16], a query) and the pair (f j ; c j ) a term of the decision list. 2.2 Threshold functions and threshold decision lists A function t : R f0; 1g is a threshold function if there are w 2 R and 2 R such that t(x) 1 if hw; xi 0 if hw; xi , where hw; xi is the standard inner ....
M. Krause. On the Computational Power of Boolean Decision Lists. In Proceedings of the 19th Annual Symposium of Theoretical Aspects of Computer Science (STACS), 2002.
Decision Lists and Threshold Decision Lists - Martin Anthony December (Correct)
....the set of finite sequences f = f 1 , c 1 ) f 2 , c 2 ) f r , c r ) such that f i K and c i 1 for 1 r. The values of f are defined by f(y) c j where j = min i f i (y) 1 , or 0 if there are no j such that f j (y) 1. We call each f j a test (or, following Krause [15], a query) and the pair (f j , c j ) a term of the decision list. Decision lists were introduced by Rivest [19] where a key concern was develop a learning algorithm for them. This is discussed later in this report. Note that we do not always draw a strong distinction between a decision list ....
....most k literals, so each test is a simple conjunction of degree at most k. Then, following Rivest [19] DL(K) is usually denoted k DL and we call such decision lists k decision lists. It should be noted, however, that this terminology has also been used to mean something slightly different: Krause [15] defines a k decision list to be one in which each test involves at most k variables, and this is a more general class than k DL. Note that when K = M n,k , we have K # and hence k DL DL(M n,k ) e = 2 O(n log n) for fixed k. Later, we will want to consider K ....
[Article contains additional citation context not shown here]
Matthias Krause. On the Computational Power of Boolean Decision Lists. In Proceedings of the 19th Annual Symposium of Theoretical Aspects of Computer Science (STACS), 2002.
Toward Attribute Efficient Learning of Decision Lists and.. - Klivans, Servedio (2006) (Correct)
No context found.
M. Krause. On the computational power of Boolean decision lists. In 19th Annual Symposium on Theoretical Aspects of Computer Science, pages 372--383, 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC