Results 11  20
of
24
BlackBox Reductions for CostSharing Mechanism Design
"... We consider the design of strategyproof costsharing mechanisms. We give two simple, but extremely versatile, blackbox reductions, that in combination reduce the costsharing mechanismdesign problem to the algorithmic problem of finding a minimumcost solution for a set of players. Our first reduc ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
We consider the design of strategyproof costsharing mechanisms. We give two simple, but extremely versatile, blackbox reductions, that in combination reduce the costsharing mechanismdesign problem to the algorithmic problem of finding a minimumcost solution for a set of players. Our first reduction shows that any truthful, αapproximation mechanism for the socialcost minimization (SCM) problem satisfying a technical nobossiness condition can be morphed into a truthful mechanism that achieves an O(α log n)approximation where the prices recover the cost incurred. Thus, we decouple the task of truthfully computing an outcome with nearoptimal social cost from the costsharing problem. This is fruitful since truthful mechanismdesign, especially for singledimensional problems, is a relatively wellunderstood and manageable task. Our second reduction nicely complements the first one by showing that any LPbased ρapproximation for the problem of finding a mincost solution for a set of players yields a truthful, nobossy, (ρ + 1)approximation for the SCM problem (and hence, a truthful (ρ + 1) log napproximation costsharing mechanism). These reductions find a slew of applications, yielding, as corollaries, the first or improved polytime costsharing mechanisms for a variety of problems. For example, our first reduction coupled with the celebrated VCG mechanism shows that for any costsharing problem (with a monotone cost function) one can obtain a truthful mechanism that achieves an O(log n)approximation where the prices recover the cost incurred. Other applications include O(log n)approximation mechanisms for: survivable network design problems, facility location (FL) problems including capacitated and connected FL problems, and minimummakespan scheduling on unrelated machines. Our results demonstrate that in contrast with our current understanding of groupstrategyproof and acyclic mechanisms, strategyproofness allows for ample flexibility in costsharing mechanism design enabling one to effectively leverage various algorithmic results.
Primaldual algorithms for combinatorial optimization problems
, 2007
"... Combinatorial optimization problems such as routing, scheduling, covering and packing problems abound in everyday life. At a very high level, a combinatorial optimization problem amounts to finding a solution with minimum or maximum cost among a large number of feasible solutions. An algorithm for a ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Combinatorial optimization problems such as routing, scheduling, covering and packing problems abound in everyday life. At a very high level, a combinatorial optimization problem amounts to finding a solution with minimum or maximum cost among a large number of feasible solutions. An algorithm for a given optimization problem is said to be exact if it always returns an optimal solution and is said to be efficient if it runs in time polynomial on the size of its input. The theory of NPcompleteness suggests that exact and efficient algorithms are unlikely to exist for the class of NPhard problems. Unfortunately, a large number of natural and interesting combinatorial optimization problems are NPhard. One way to cope with NPhardness is to relax the optimality requirement and instead look for solutions that are provably close to the optimum. This is the main idea behind approximation algorithms. An algorithm is said to be a ρapproximation if it always returns a solution whose cost is at most a ρ factor away from the optimal cost. Arguably, one of the most important techniques in the design of combinatorial algorithms is the primaldual schema in which the cost of the primal solution is compared to the cost of a dual solution. In this dissertation we study the primaldual schema in the design of approximation algorithms for a number of covering and scheduling problems.
LAGRANGIAN RELAXATION AND PARTIAL COVER
"... Lagrangian relaxation has been used extensively in the design of approximation algorithms. This paper studies its strengths and limitations when applied to Partial Cover. We show that for Partial Cover in general no algorithm that uses Lagrangian relaxation and a Lagrangian Multiplier Preserving (LM ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Lagrangian relaxation has been used extensively in the design of approximation algorithms. This paper studies its strengths and limitations when applied to Partial Cover. We show that for Partial Cover in general no algorithm that uses Lagrangian relaxation and a Lagrangian Multiplier Preserving (LMP) αapproximation as a black box can yield an approximation factor better than 4/3 α. This matches the upper bound given by Könemann et al. (ESA 2006, pages 468–479). Faced with this limitation we study a specific, yet broad class of covering problems: Partial Totally Balanced Cover. By carefully analyzing the inner workings of the LMP algorithm we are able to give an almost tight characterization of the integrality gap of the standard linear relaxation of the problem. As a consequence we obtain improved approximations for the Partial version of Multicut and Path Hitting on Trees, Rectangle Stabbing, and Set Cover with ρBlocks.
Partial Interval Set Cover  Tradeoffs Between Scalability and Optimality
, 2013
"... Given an interval I = {1, 2,..., n} of points, a collection I of subintervals of I and a fraction 0 ≤ r ≤ 1, we consider the following variation of partial set cover. We wish to find an optimal subset of I covering at least an rfraction of I. While this problem is easily solved exactly in quadratic ..."
Abstract
 Add to MetaCart
(Show Context)
Given an interval I = {1, 2,..., n} of points, a collection I of subintervals of I and a fraction 0 ≤ r ≤ 1, we consider the following variation of partial set cover. We wish to find an optimal subset of I covering at least an rfraction of I. While this problem is easily solved exactly in quadratic time using classical methods, we focus on developing scalable algorithms which return nearoptimal solutions and run in nearlinear time. We give a (1 + ɛ)approximation algorithm running in O ( 1 ɛ · min{n + I, I  log I}}) time. We also prove a tight approximation ratio of 2 for a simple greedy algorithm for this problem, improving on the bound of 9 given in [10].
Partial Multicovering and the dconsecutive Ones Property
, 2011
"... A dinterval is the union of d disjoint intervals on the real line. In the dinterval stabbing problem (dis) we are given a set of dintervals and a set of points, each dinterval I has a stabbing requirement r(I) and each point has a weight, and the goal is to find a minimum weight multiset of poi ..."
Abstract
 Add to MetaCart
(Show Context)
A dinterval is the union of d disjoint intervals on the real line. In the dinterval stabbing problem (dis) we are given a set of dintervals and a set of points, each dinterval I has a stabbing requirement r(I) and each point has a weight, and the goal is to find a minimum weight multiset of points that stabs each dinterval I at least r(I) times. In practice there is a tradeoff between fulfilling requirements and cost, and therefore it is interesting to study problems in which one is required to fulfill only a subset of the requirements. In this paper we study variants of dis in which a feasible solution is a multiset of points that may satisfy only a subset of the stabbing requirements. In partial dis we are given an integer t, and the sum of requirements satisfied by the computed solution must be at least t. In prize collecting dis each dinterval has a penalty that must be paid for every unit of unsatisfied requirement. We also consider a maximization version of prize collecting dis in which each dinterval has a prize that is gained for every time, up to r(I), it is stabbed. Our study is motivated by several resource allocation and geometric facility location problems. We present a (ρ+d−1 ρ)approximation algorithm for prize collecting dis, where ρ = minI r(I), and an O(d)approximation algorithm for partial dis. We obtain the latter result by designing a general framework for approximating partial multicovering problems that extends the framework for approximating partial covering problems from [21]. We also show that maximum prize collecting dis is at least as hard to approximate as maximum independent set, even for d = 2, and present a dapproximation algorithm for maximum prize collecting ddimensional rectangle stabbing.
Classification by Set Cover: The Prototype Vector Machine
, 2009
"... We introduce a new nearestprototype classifier, the prototype vector machine (PVM). It arises from a combinatorial optimization problem which we cast as a variant of the set cover problem. We propose two algorithms for approximating its solution. The PVM selects a relatively small number of represe ..."
Abstract
 Add to MetaCart
(Show Context)
We introduce a new nearestprototype classifier, the prototype vector machine (PVM). It arises from a combinatorial optimization problem which we cast as a variant of the set cover problem. We propose two algorithms for approximating its solution. The PVM selects a relatively small number of representative points which can then be used for classification. It contains 1NN as a special case. The method is compatible with any dissimilarity measure, making it amenable to situations in which the data are not embedded in an underlying feature space or in which using a nonEuclidean metric is desirable. Indeed, we demonstrate on the much studied ZIP code data how the PVM can reap the benefits of a problemspecific metric. In this example, the PVM outperforms the highly successful 1NN with tangent distance, and does so retaining fewer than half of the data points. This example highlights the strengths of the PVM in yielding a lowerror, highly interpretable model. Additionally, we apply the PVM to a protein classification problem in which a kernelbased distance is used. 1
FPTAS’s for Trimming Weighted Trees
"... Given a tree with nonnegative edge cost and nonnegative vertex weight, and a number k ≥ 0, we consider the following four cut problems: cutting vertices of weight at most or at least k from the tree by deleting some edges such that the remaining part of the graph is still a tree and the total cost o ..."
Abstract
 Add to MetaCart
(Show Context)
Given a tree with nonnegative edge cost and nonnegative vertex weight, and a number k ≥ 0, we consider the following four cut problems: cutting vertices of weight at most or at least k from the tree by deleting some edges such that the remaining part of the graph is still a tree and the total cost of the edges being deleted is minimized or maximized. The MinMstCut problem (cut vertices of weight at most k and minimize the total cost of the edges being deleted) can be solved in linear time and space and the other three problems are NPhard. In this paper, we design an O(nl/ε)time O(l 2 /ε + n)space algorithm for MaxMstCut, and O(nl(1/ε+log n))time O(l 2 /ε+n)space algorithms for the other two problems, MinLstCut and MaxLstCut, where n is the number of vertices in the tree, l the number of leaves, and ε> 0 the prescribed error bound. Key words.
BlackBox Reductions for CostSharing Mechanism
"... We consider the design of strategyproof costsharing mechanisms. We give two simple, but extremely versatile, blackbox reductions, that in combination reduce the costsharing mechanismdesign problem to the algorithmic problem of finding a mincost solution for a set of players. Our first reductio ..."
Abstract
 Add to MetaCart
We consider the design of strategyproof costsharing mechanisms. We give two simple, but extremely versatile, blackbox reductions, that in combination reduce the costsharing mechanismdesign problem to the algorithmic problem of finding a mincost solution for a set of players. Our first reduction shows that any truthful, αapproximation mechanism for the socialcost minimization (SCM) problem satisfying a technical nobossiness condition can be morphed into a truthful mechanism that achieves an O(α log n)approximation where the prices recover the cost incurred. Thus, we decouple (modulo nobossiness) the task of truthfully computing an outcome with nearoptimal social cost from the costsharing problem. This is fruitful since truthful mechanismdesign, especially for singledimensional problems, is a relatively wellunderstood and manageable task. Our second reduction nicely complements the first one by showing that any LPrelative ρapproximation for the problem of finding a mincost solution for a set of players yields a truthful, nobossy, (ρ+ 1)approximation for the SCM problem (and hence, a truthful (ρ + 1) log napproximation costsharing mechanism).