arXiv subjects

Alina Ene

Publications and source records attributed to Alina Ene.

50 records · Page 3Linked to original sources

arXiv2016

On Approximating Maximum Independent Set of Rectangles

We study the Maximum Independent Set of Rectangles (MISR) problem: given a set of $n$ axis-parallel rectangles, find a largest-cardinality subset of the rectangles, such that no two of them overlap. MISR is a basic geometric optimization problem with many applications, that has been studied extensively. Until recently, the best approximation algorithm for it achieved an $O(\log \log n)$-approximation factor. In a recent breakthrough, Adamaszek and Wiese provided a quasi-polynomial time approximation scheme: a $(1-ε)$-approximation algorithm with running time $n^{O(\operatorname{poly}(\log n)/ε)}$. Despite this result, obtaining a PTAS or even a polynomial-time constant-factor approximation remains a challenging open problem. In this paper we make progress towards this goal by providing an algorithm for MISR that achieves a $(1 - ε)$-approximation in time $n^{O(\operatorname{poly}(\log\log{n} / ε))}$. We introduce several new technical ideas, that we hope will lead to further progress on this and related problems.

cs.DS↗

arXiv2016

Routing under Balance

We introduce the notion of balance for directed graphs: a weighted directed graph is $α$-balanced if for every cut $S \subseteq V$, the total weight of edges going from $S$ to $V\setminus S$ is within factor $α$ of the total weight of edges going from $V\setminus S$ to $S$. Several important families of graphs are nearly balanced, in particular, Eulerian graphs (with $α= 1$) and residual graphs of $(1+ε)$-approximate undirected maximum flows (with $α=O(1/ε)$). We use the notion of balance to give a more fine-grained understanding of several well-studied routing questions that are considerably harder in directed graphs. We first revisit oblivious routings in directed graphs. Our main algorithmic result is an oblivious routing scheme for single-source instances that achieve an $O(α\cdot \log^3 n / \log \log n)$ competitive ratio. In the process, we make several technical contributions which may be of independent interest. In particular, we give an efficient algorithm for computing low-radius decompositions of directed graphs parameterized by balance. We also define and construct low-stretch arborescences, a generalization of low-stretch spanning trees to directed graphs. On the negative side, we present new lower bounds for oblivious routing problems on directed graphs. We show that the competitive ratio of oblivious routing algorithms for directed graphs is $Ω(n)$ in general; this result improves upon the long-standing best known lower bound of $Ω(\sqrt{n})$ given by Hajiaghayi, Kleinberg, Leighton and Räcke in 2006. We also show that our restriction to single-source instances is necessary by showing an $Ω(\sqrt{n})$ lower bound for multiple-source oblivious routing in Eulerian graphs. We also give a fast algorithm for the maximum flow problem in balanced directed graphs.

cs.DS↗

arXiv2015

On Routing Disjoint Paths in Bounded Treewidth Graphs

We study the problem of routing on disjoint paths in bounded treewidth graphs with both edge and node capacities. The input consists of a capacitated graph $G$ and a collection of $k$ source-destination pairs $\mathcal{M} = \{(s_1, t_1), \dots, (s_k, t_k)\}$. The goal is to maximize the number of pairs that can be routed subject to the capacities in the graph. A routing of a subset $\mathcal{M}'$ of the pairs is a collection $\mathcal{P}$ of paths such that, for each pair $(s_i, t_i) \in \mathcal{M}'$, there is a path in $\mathcal{P}$ connecting $s_i$ to $t_i$. In the Maximum Edge Disjoint Paths (MaxEDP) problem, the graph $G$ has capacities $\mathrm{cap}(e)$ on the edges and a routing $\mathcal{P}$ is feasible if each edge $e$ is in at most $\mathrm{cap}(e)$ of the paths of $\mathcal{P}$. The Maximum Node Disjoint Paths (MaxNDP) problem is the node-capacitated counterpart of MaxEDP. In this paper we obtain an $O(r^3)$ approximation for MaxEDP on graphs of treewidth at most $r$ and a matching approximation for MaxNDP on graphs of pathwidth at most $r$. Our results build on and significantly improve the work by Chekuri et al. [ICALP 2013] who obtained an $O(r \cdot 3^r)$ approximation for MaxEDP.

cs.DS↗

arXiv2015

Online Buy-at-Bulk Network Design

We present the first non-trivial online algorithms for the non-uniform, multicommodity buy-at-bulk (MC-BB) network design problem in undirected and directed graphs. Our competitive ratios qualitatively match the best known approximation factors for the corresponding offline problems. The main engine for our results is an online reduction theorem of MC-BB problems to their single-sink (SS-BB) counterparts. We use the concept of junction-tree solutions (Chekuri et al., FOCS 2006) that play an important role in solving the offline versions of the problem via a greedy subroutine -- an inherently offline procedure. Our main technical contribution is in designing an online algorithm using only the existence of good junction-trees to reduce an MC-BB instance to multiple SS-BB sub-instances. Along the way, we also give the first non-trivial online node-weighted/directed single-sink buy-at-bulk algorithms. In addition to the new results, our generic reduction also yields new proofs of recent results for the online node-weighted Steiner forest and online group Steiner forest problems.

cs.DS↗

arXiv2015

The Power of Randomization: Distributed Submodular Maximization on Massive Datasets

A wide variety of problems in machine learning, including exemplar clustering, document summarization, and sensor placement, can be cast as constrained submodular maximization problems. Unfortunately, the resulting submodular optimization problems are often too large to be solved on a single machine. We develop a simple distributed algorithm that is embarrassingly parallel and it achieves provable, constant factor, worst-case approximation guarantees. In our experiments, we demonstrate its efficiency in large problems with different kinds of constraints with objective values always close to what is achievable in the centralized setting.

cs.LG↗

arXiv2015

Local Distribution and the Symmetry Gap: Approximability of Multiway Partitioning Problems

We study the approximability of multiway partitioning problems, examples of which include Multiway Cut, Node-weighted Multiway Cut, and Hypergraph Multiway Cut. We investigate these problems from the point of view of two possible generalizations: as Min-CSPs, and as Submodular Multiway Partition problems. These two generalizations lead to two natural relaxations, the Basic LP, and the Lovasz relaxation. We show that the Lovasz relaxation gives a (2-2/k)-approximation for Submodular Multiway Partition with $k$ terminals, improving a recent 2-approximation. We prove that this factor is optimal in two senses: (1) A (2-2/k-ε)-approximation for Submodular Multiway Partition with k terminals would require exponentially many value queries. (2) For Hypergraph Multiway Cut and Node-weighted Multiway Cut with k terminals, both special cases of Submodular Multiway Partition, we prove that a (2-2/k-ε)-approximation is NP-hard, assuming the Unique Games Conjecture. Both our hardness results are more general: (1) We show that the notion of symmetry gap, previously used for submodular maximization problems, also implies hardness results for submodular minimization problems. (2) Assuming the Unique Games Conjecture, we show that the Basic LP gives an optimal approximation for every Min-CSP that includes the Not-Equal predicate. Finally, we connect the two hardness techniques by proving that the integrality gap of the Basic LP coincides with the symmetry gap of the multilinear relaxation (for a related instance). This shows that the appearance of the same hardness threshold for a Min-CSP and the related submodular minimization problem is not a coincidence.

cs.DS↗

arXiv2015

Random Coordinate Descent Methods for Minimizing Decomposable Submodular Functions

Submodular function minimization is a fundamental optimization problem that arises in several applications in machine learning and computer vision. The problem is known to be solvable in polynomial time, but general purpose algorithms have high running times and are unsuitable for large-scale problems. Recent work have used convex optimization techniques to obtain very practical algorithms for minimizing functions that are sums of ``simple" functions. In this paper, we use random coordinate descent methods to obtain algorithms with faster linear convergence rates and cheaper iteration costs. Compared to alternating projection methods, our algorithms do not rely on full-dimensional vector operations and they converge in significantly fewer iterations.

cs.LG↗

arXiv2013

Connected Domatic Packings in Node-capacitated Graphs

A set of vertices in a graph is a dominating set if every vertex outside the set has a neighbor in the set. A dominating set is connected if the subgraph induced by its vertices is connected. The connected domatic partition problem asks for a partition of the nodes into connected dominating sets. The connected domatic number of a graph is the size of a largest connected domatic partition and it is a well-studied graph parameter with applications in the design of wireless networks. In this note, we consider the fractional counterpart of the connected domatic partition problem in \emph{node-capacitated} graphs. Let $n$ be the number of nodes in the graph and let $k$ be the minimum capacity of a node separator in $G$. Fractionally we can pack at most $k$ connected dominating sets subject to the capacities on the nodes, and our algorithms construct packings whose sizes are proportional to $k$. Some of our main contributions are the following: \begin{itemize} \item An algorithm for constructing a fractional connected domatic packing of size $Ω(k)$ for node-capacitated planar and minor-closed families of graphs. \item An algorithm for constructing a fractional connected domatic packing of size $Ω(k / \ln{n})$ for node-capacitated general graphs. \end{itemize}

cs.DS↗

arXiv2013

Fast Clustering with Lower Bounds: No Customer too Far, No Shop too Small

We study the \LowerBoundedCenter (\lbc) problem, which is a clustering problem that can be viewed as a variant of the \kCenter problem. In the \lbc problem, we are given a set of points P in a metric space and a lower bound λ, and the goal is to select a set C \subseteq P of centers and an assignment that maps each point in P to a center of C such that each center of C is assigned at least λpoints. The price of an assignment is the maximum distance between a point and the center it is assigned to, and the goal is to find a set of centers and an assignment of minimum price. We give a constant factor approximation algorithm for the \lbc problem that runs in O(n \log n) time when the input points lie in the d-dimensional Euclidean space R^d, where d is a constant. We also prove that this problem cannot be approximated within a factor of 1.8-εunless P = \NP even if the input points are points in the Euclidean plane R^2.

cs.CG↗

arXiv2011

Geometric Packing under Non-uniform Constraints

We study the problem of discrete geometric packing. Here, given weighted regions (say in the plane) and points (with capacities), one has to pick a maximum weight subset of the regions such that no point is covered more than its capacity. We provide a general framework and an algorithm for approximating the optimal solution for packing in hypergraphs arising out of such geometric settings. Using this framework we get a flotilla of results on this problem (and also on its dual, where one wants to pick a maximum weight subset of the points when the regions have capacities). For example, for the case of fat triangles of similar size, we show an O(1)-approximation and prove that no \PTAS is possible.

cs.CG↗

arXiv2011

Fast Clustering using MapReduce

Clustering problems have numerous applications and are becoming more challenging as the size of the data increases. In this paper, we consider designing clustering algorithms that can be used in MapReduce, the most popular programming environment for processing large datasets. We focus on the practical and popular clustering problems, $k$-center and $k$-median. We develop fast clustering algorithms with constant factor approximation guarantees. From a theoretical perspective, we give the first analysis that shows several clustering algorithms are in $\mathcal{MRC}^0$, a theoretical MapReduce class introduced by Karloff et al. \cite{KarloffSV10}. Our algorithms use sampling to decrease the data size and they run a time consuming clustering algorithm such as local search or Lloyd's algorithm on the resulting data set. Our algorithms have sufficient flexibility to be used in practice since they run in a constant number of MapReduce rounds. We complement these results by performing experiments using our algorithms. We compare the empirical performance of our algorithms to several sequential and parallel algorithms for the $k$-median problem. The experiments show that our algorithms' solutions are similar to or better than the other algorithms' solutions. Furthermore, on data sets that are sufficiently large, our algorithms are faster than the other parallel algorithms that we tested.

cs.DC↗

arXiv2011

Submodular Cost Allocation Problem and Applications

We study the Minimum Submodular-Cost Allocation problem (MSCA). In this problem we are given a finite ground set $V$ and $k$ non-negative submodular set functions $f_1 ,..., f_k$ on $V$. The objective is to partition $V$ into $k$ (possibly empty) sets $A_1 ,..., A_k$ such that the sum $\sum_{i=1}^k f_i(A_i)$ is minimized. Several well-studied problems such as the non-metric facility location problem, multiway-cut in graphs and hypergraphs, and uniform metric labeling and its generalizations can be shown to be special cases of MSCA. In this paper we consider a convex-programming relaxation obtained via the Lovász-extension for submodular functions. This allows us to understand several previous relaxations and rounding procedures in a unified fashion and also develop new formulations and approximation algorithms for several problems. In particular, we give a $(1.5 - 1/k)$-approximation for the hypergraph multiway partition problem. We also give a $\min\{2(1-1/k), H_Δ\}$-approximation for the hypergraph multiway cut problem when $Δ$ is the maximum hyperedge size. Both problems generalize the multiway cut problem in graphs and the hypergraph cut problem is approximation equivalent to the node-weighted multiway cut problem in graphs.

cs.DS↗

arXiv2011

Approximation Algorithms for Submodular Multiway Partition

We study algorithms for the Submodular Multiway Partition problem (SubMP). An instance of SubMP consists of a finite ground set $V$, a subset of $k$ elements $S = \{s_1,s_2,...,s_k\}$ called terminals, and a non-negative submodular set function $f:2^V\rightarrow \mathbb{R}_+$ on $V$ provided as a value oracle. The goal is to partition $V$ into $k$ sets $A_1,...,A_k$ such that for $1 \le i \le k$, $s_i \in A_i$ and $\sum_{i=1}^k f(A_i)$ is minimized. SubMP generalizes some well-known problems such as the Multiway Cut problem in graphs and hypergraphs, and the Node-weighed Multiway Cut problem in graphs. SubMP for arbitrarysubmodular functions (instead of just symmetric functions) was considered by Zhao, Nagamochi and Ibaraki \cite{ZhaoNI05}. Previous algorithms were based on greedy splitting and divide and conquer strategies. In very recent work \cite{ChekuriE11} we proposed a convex-programming relaxation for SubMP based on the Lovász-extension of a submodular function and showed its applicability for some special cases. In this paper we obtain the following results for arbitrary submodular functions via this relaxation. (i) A 2-approximation for SubMP. This improves the $(k-1)$-approximation from \cite{ZhaoNI05} and (ii) A $(1.5-1/k)$-approximation for SubMP when $f$ is symmetric. This improves the $2(1-1/k)$-approximation from \cite{Queyranne99,ZhaoNI05}.

cs.DS↗

arXiv2010

Prize-Collecting Steiner Tree and Forest in Planar Graphs

We obtain polynomial-time approximation-preserving reductions (up to a factor of 1 + ε) from the prize-collecting Steiner tree and prize-collecting Steiner forest problems in planar graphs to the corresponding problems in graphs of bounded treewidth. We also give an exact algorithm for the prize-collecting Steiner tree problem that runs in polynomial time for graphs of bounded treewidth. This, combined with our reductions, yields a PTAS for the prize-collecting Steiner tree problem in planar graphs and generalizes the PTAS of Borradaile, Klein and Mathieu for the Steiner tree problem in planar graphs. Our results build upon the ideas of Borradaile, Klein and Mathieu and the work of Bateni, Hajiaghayi and Marx on a PTAS for the Steiner forest problem in planar graphs. Our main technical result is on the properties of primal-dual algorithms for Steiner tree and forest problems in general graphs when they are run with scaled up penalties.

cs.DS↗