Decision list compression by mild random restrictions

Decision list compression by mild random restrictions
复制标题

通过轻度随机限制进行决策列表压缩

DOI:
10.1145/3357713.3384241
复制
发表时间:
2020
期刊:
Proceedings of the 52nd Annual ACM SIGACT Symposium on Theory of Computing (STOC 2020
影响因子:
--
通讯作者:
Zhang, Jiapeng
Zhang, Jiapeng
中科院分区:
--
文献类型:
--
作者:
Lovett, Shachar;Wu, Kewen;Zhang, Jiapeng

文献摘要

参考文献

被引文献

相似文献

决策列表是规则的有序列表。每个规则都由一个项和一个值指定,该项是文字的合取。给定一个输入,决策列表的输出是对应于第一个规则的值,该规则的项由输入满足。决策表是CNFs和DNF的推广,在复杂性理论和学习理论中都有研究,决策表的大小是规则的数量,宽度是一个术语中变量的最大数量。我们证明了小宽度的决策列表总是可以近似的决策列表的小尺寸,在那里我们得到这样的近似的尖锐的界限。这也解决了Gopalan,Meka和Reingold(Computational Complexity,2013)关于DNF稀疏化的猜想。我们证明中的一个成分是一个新的随机限制引理,它允许分析如果一小部分变量是固定的,DNF(更一般地说,决策列表)如何简化。这与更常用的开关引理相反,开关引理要求大多数变量是固定的。
A decision list is an ordered list of rules. Each rule is specified by a term, which is a conjunction of literals, and a value. Given an input, the output of a decision list is the value corresponding to the first rule whose term is satisfied by the input. Decision lists generalize both CNFs and DNFs and have been studied both in complexity theory and in learning theory.The size of a decision list is the number of rules, and its width is the maximal number of variables in a term. We prove that decision lists of small width can always be approximated by decision lists of small size, where we obtain sharp bounds for such approximation. This also resolves a conjecture of Gopalan, Meka, and Reingold (Computational Complexity, 2013) on DNF sparsification.An ingredient in our proof is a new random restriction lemma, which allows to analyze how DNFs (and more generally, decision lists) simplify if a small fraction of the variables are fixed. This is in contrast to the more commonly used switching lemma, which requires most of the variables to be fixed.
DOI: --
发表时间: 1997
期刊:
影响因子: --
作者:
György Turán;F. Vatan
通讯作者: F. Vatan
DOI: --
发表时间: 2014
期刊: Journal of management science
影响因子: --
作者:
อนิรุธ สืบสิงห์
通讯作者: อนิรุธ สืบสิงห์
论决策表的在线学习
DOI: --
发表时间: 2003
影响因子: 6
作者:
Ziv Nevo;Ran El
通讯作者: Ran El
论布尔公式的可学习性
DOI: --
发表时间: 1987
期刊: Symposium on the Theory of Computing
影响因子: --
作者:
M. Kearns;Ming Li;L. Pitt;L. Valiant
通讯作者: L. Valiant
关于决策列表的研究笔记
DOI: --
发表时间: 1993
期刊: Machine-mediated learning
影响因子: --
作者:
Ron Kohavi;S. Benson
通讯作者: S. Benson