All classical adversary methods are equivalent for total functions
From MaRDI portal
Abstract: We show that all known classical adversary lower bounds on randomized query complexity are equivalent for total functions, and are equal to the fractional block sensitivity . That includes the Kolmogorov complexity bound of Laplante and Magniez and the earlier relational adversary bound of Aaronson. This equivalence also implies that for total functions, the relational adversary is equivalent to a simpler lower bound, which we call rank-1 relational adversary. For partial functions, we show unbounded separations between and other adversary bounds, as well as between the adversary bounds themselves. We also show that, for partial functions, fractional block sensitivity cannot give lower bounds larger than , where is the number of variables and is the block sensitivity. Then we exhibit a partial function that matches this upper bound, .
Recommendations
Cites work
- An introduction to Kolmogorov complexity and its applications
- Complexity measures and decision tree complexity: a survey.
- Composition limits and separating examples for some Boolean function complexity measures
- scientific article; zbMATH DE number 5899240 (Why is no real title available?)
- scientific article; zbMATH DE number 5485488 (Why is no real title available?)
- Lower Bounds for Local Search by Quantum Arguments
- On fractional block sensitivity
- On the degree of Boolean functions as real polynomials
- Properties and applications of Boolean function composition
- Quantum certificate complexity
- Quantum lower bounds by polynomials
- Quantum lower bounds by quantum arguments
- Randomized query complexity of sabotaged and composed functions
- Separations in query complexity using cheat sheets
- Span Programs and Quantum Query Complexity: The General Adversary Bound Is Nearly Tight for Every Boolean Function
Cited in
(4)
This page was built for publication: All classical adversary methods are equivalent for total functions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3304102)