From 411e59045922c4d50d14fb30aa5e0bdeecf42991 Mon Sep 17 00:00:00 2001 From: Stratis Ioannidis Date: Sat, 6 Jul 2013 14:05:17 -0700 Subject: intro sufficiently concave --- abstract.tex | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'abstract.tex') diff --git a/abstract.tex b/abstract.tex index d62c1d5..2f33699 100644 --- a/abstract.tex +++ b/abstract.tex @@ -19,6 +19,6 @@ Each subject $i$ declares an associated cost $c_i >0$ to be part of the experime mechanism for \SEDP{} with suitable properties. We present a deterministic, polynomial time, $\delta$-truthful, budget feasible mechanism for \SEDP{}. -By applying previous work on budget feasible mechanisms with submodular objective, one could {\em only} have derived either an exponential time deterministic mechanism or a randomized polynomial time mechanism. Our mechanism yields a constant factor ($\approx 12.68$) approximation, and we show that no truthful, budget-feasible algorithms are possible within a factor $2$ approximation. We also show how to generalize our approach to a wide class of learning problems. +By applying previous work on budget feasible mechanisms with submodular objective, one could {\em only} have derived either an exponential time deterministic mechanism or a randomized polynomial time mechanism. Our mechanism yields a constant factor ($\approx 12.68$) approximation, and we show that no truthful, budget-feasible algorithms are possible within a factor $2$ approximation. We also show how to generalize our approach to a wide class of learning problems, beyond linear regression. -- cgit v1.2.3-70-g09d2