More Like this

1. Parameterized Approximation Scheme for Feedback Vertex Set

   https://doi.org/10.4230/LIPIcs.MFCS.2023.56  

   Jana, Satyabrata; Lokshtanov, Daniel; Mandal, Soumen; Rai, Ashutosh; Saurabh, Saket (January 2023, Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Leroux, Jérôme; Lombardy, Sylvain; Peleg, David (Ed.)

   Feedback Vertex Set (FVS) is one of the most studied vertex deletion problems in the field of graph algorithms. In the decision version of the problem, given a graph G and an integer k, the question is whether there exists a set S of at most k vertices in G such that G-S is acyclic. It is one of the first few problems which were shown to be NP-complete, and has been extensively studied from the viewpoint of approximation and parameterized algorithms. The best-known polynomial time approximation algorithm for FVS is a 2-factor approximation, while the best known deterministic and randomized FPT algorithms run in time 𝒪^*(3.460^k) and 𝒪^*(2.7^k) respectively. In this paper, we contribute to the newly established area of parameterized approximation, by studying FVS in this paradigm. In particular, we combine the approaches of parameterized and approximation algorithms for the study of FVS, and achieve an approximation guarantee with a factor better than 2 in randomized FPT running time, that improves over the best known parameterized algorithm for FVS. We give three simple randomized (1+ε) approximation algorithms for FVS, running in times 𝒪^*(2^{εk}⋅ 2.7^{(1-ε)k}), 𝒪^*(({(4/(1+ε))^{(1+ε)}}⋅{(ε/3)^ε})^k), and 𝒪^*(4^{(1-ε)k}) respectively for every ε ∈ (0,1). Combining these three algorithms, we obtain a factor (1+ε) approximation algorithm for FVS, which has better running time than the best-known (randomized) FPT algorithm for every ε ∈ (0, 1). This is the first attempt to look at a parameterized approximation of FVS to the best of our knowledge. Our algorithms are very simple, and they rely on some well-known reduction rules used for arriving at FPT algorithms for FVS. 

   more » « less
2. Two-Dimensional Longest Common Extension Queries in Compact Space

   https://doi.org/10.4230/LIPICS.STACS.2025.38  

   Ganguly, Arnab; Gibney, Daniel; Shah, Rahul; Thankachan, Sharma V (January 2025, Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Beyersdorff, Olaf; Pilipczuk, Michał; Pimentel, Elaine; Thắng, Nguyễn Kim (Ed.)

   For a length n text over an alphabet of size σ, we can encode the suffix tree data structure in 𝒪(nlog σ) bits of space. It supports suffix array (SA), inverse suffix array (ISA), and longest common extension (LCE) queries in 𝒪(log^ε_σ n) time, which enables efficient pattern matching; here ε > 0 is an arbitrarily small constant. Further improvements are possible for LCE queries, where 𝒪(1) time queries can be achieved using an index of space 𝒪(nlog σ) bits. However, compactly indexing a two-dimensional text (i.e., an n× n matrix) has been a major open problem. We show progress in this direction by first presenting an 𝒪(n²log σ)-bit structure supporting LCE queries in near 𝒪((log_σ n)^{2/3}) time. We then present an 𝒪(n²log σ + n²log log n)-bit structure supporting ISA queries in near 𝒪(log n ⋅ (log_σ n)^{2/3}) time. Within a similar space, achieving SA queries in poly-logarithmic (even strongly sub-linear) time is a significant challenge. However, our 𝒪(n²log σ + n²log log n)-bit structure can support SA queries in 𝒪(n²/(σ log n)^c) time, where c is an arbitrarily large constant, which enables pattern matching in time faster than what is possible without preprocessing. We then design a repetition-aware data structure. The δ_2D compressibility measure for two-dimensional texts was recently introduced by Carfagna and Manzini [SPIRE 2023]. The measure ranges from 1 to n², with smaller δ_2D indicating a highly compressible two-dimensional text. The current data structure utilizing δ_2D allows only element access. We obtain the first structure based on δ_2D for LCE queries. It takes 𝒪^{~}(n^{5/3} + n^{8/5}δ_2D^{1/5}) space and answers queries in 𝒪(log n) time. 

   more » « less
3. Breaking the All Subsets Barrier for Min k-Cut

   https://doi.org/10.4230/LIPIcs.ICALP.2023.90  

   Lokshtanov, Daniel; Saurabh, Saket; Surianarayanan, Vaishali (January 2023, Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Etessami, Kousha; Feige, Uriel; Puppis, Gabriele (Ed.)

   In the Min k-Cut problem, the input is a graph G and an integer k. The task is to find a partition of the vertex set of G into k parts, while minimizing the number of edges that go between different parts of the partition. The problem is NP-complete, and admits a simple 3ⁿ⋅n^𝒪(1) time dynamic programming algorithm, which can be improved to a 2ⁿ⋅n^𝒪(1) time algorithm using the fast subset convolution framework by Björklund et al. [STOC'07]. In this paper we give an algorithm for Min k-Cut with running time 𝒪((2-ε)ⁿ), for ε > 10^{-50}. This is the first algorithm for Min k-Cut with running time 𝒪(cⁿ) for c < 2. 

   more » « less
4. One-Way Communication Complexity of Minimum Vertex Cover in General Graphs

   https://doi.org/10.4230/lipics.icalp.2025.66  

   Derakhshan, Mahsa; Ghasemi, Andisheh; Rajaraman, Rajmohan (January 2025, Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Censor-Hillel, Keren; Grandoni, Fabrizio; Ouaknine, Joel; Puppis, Gabriele (Ed.)

   We study the communication complexity of the Minimum Vertex Cover (MVC) problem on general graphs within the k-party one-way communication model. Edges of an arbitrary n-vertex graph are distributed among k parties. The objective is for the parties to collectively find a small vertex cover of the graph while adhering to a communication protocol where each party sequentially sends a message to the next until the last party outputs a valid vertex cover of the whole graph. We are particularly interested in the trade-off between the size of the messages sent and the approximation ratio of the output solution. It is straightforward to see that any constant approximation protocol for MVC requires communicating Ω(n) bits. Additionally, there exists a trivial 2-approximation protocol where the parties collectively find a maximal matching of the graph greedily and return the subset of vertices matched. This raises a natural question: What is the best approximation ratio achievable using optimal communication of O(n)? We design a protocol with an approximation ratio of (2-2^{-k+1}+ε) and O(n) communication for any desirably small constant ε > 0, which is strictly better than 2 for any constant number of parties. Moreover, we show that achieving an approximation ratio smaller than 3/2 for the two-party case requires n^{1 + Ω(1/lg lg n)} communication, thereby establishing the tightness of our protocol for two parties. A notable aspect of our protocol is that no edges are communicated between the parties. Instead, for any 1 ≤ i < k, the i-th party only communicates a constant number of vertex covers for all edges assigned to the first i parties. An interesting consequence is that the communication cost of our protocol is O(n) bits, as opposed to the typical Ω(nlog n) bits required for many graph problems, such as maximum matching, where protocols commonly involve communicating edges. 

   more » « less
5. Bit Complexity of Jordan Normal Form and Polynomial Spectral Factorization

   Dey, Papri; Kannan, Ravi; Ryder, Nick; Srivastava, Nikhil (January 2023, 14th Innovations in Theoretical Computer Science Conference (ITCS 2023))

   We study the bit complexity of two related fundamental computational problems in linear algebra and control theory. Our results are: (1) An Õ(n^{ω+3}a+n⁴a²+n^ωlog(1/ε)) time algorithm for finding an ε-approximation to the Jordan Normal form of an integer matrix with a-bit entries, where ω is the exponent of matrix multiplication. (2) An Õ(n⁶d⁶a+n⁴d⁴a²+n³d³log(1/ε)) time algorithm for ε-approximately computing the spectral factorization P(x) = Q^*(x)Q(x) of a given monic n× n rational matrix polynomial of degree 2d with rational a-bit coefficients having a-bit common denominators, which satisfies P(x)⪰0 for all real x. The first algorithm is used as a subroutine in the second one. Despite its being of central importance, polynomial complexity bounds were not previously known for spectral factorization, and for Jordan form the best previous best running time was an unspecified polynomial in n of degree at least twelve [Cai, 1994]. Our algorithms are simple and judiciously combine techniques from numerical and symbolic computation, yielding significant advantages over either approach by itself. 

   more » « less