arXiv subjects

Xiaorui Sun

Publications and source records attributed to Xiaorui Sun.

At least 19 recordsLinked to original sources

arXiv2024

On Efficient Variants of Segment Anything Model: A Survey

The Segment Anything Model (SAM) is a foundational model for image segmentation tasks, known for its strong generalization across diverse applications. However, its impressive performance comes with significant computational and resource demands, making it challenging to deploy in resource-limited environments such as edge devices. To address this, a variety of SAM variants have been proposed to enhance efficiency while keeping accuracy. This survey provides the first comprehensive review of these efficient SAM variants. We begin by exploring the motivations driving this research. We then present core techniques used in SAM and model acceleration. This is followed by a detailed exploration of SAM acceleration strategies, categorized by approach, and a discussion of several future research directions. Finally, we offer a unified and extensive evaluation of these methods across various hardware, assessing their efficiency and accuracy on representative benchmarks, and providing a clear comparison of their overall performance.

cs.CV↗

arXiv2024

Canonical forms for matrix tuples in polynomial time

Left-right and conjugation actions on matrix tuples have received considerable attention in theoretical computer science due to their connections with polynomial identity testing, group isomorphism, and tensor isomorphism. In this paper, we present polynomial-time algorithms for computing canonical forms of matrix tuples over a finite field under these actions. Our algorithm builds upon new structural insights for matrix tuples, which can be viewed as a generalization of Schur's lemma for irreducible representations to general representations.

cs.DS↗

arXiv2024

Fully Dynamic Min-Cut of Superconstant Size in Subpolynomial Time

We present a deterministic fully dynamic algorithm with subpolynomial worst-case time per graph update such that after processing each update of the graph, the algorithm outputs a minimum cut of the graph if the graph has a cut of size at most $c$ for some $c = (\log n)^{o(1)}$. Previously, the best update time was $\widetilde O(\sqrt{n})$ for any $c > 2$ and $c = O(\log n)$ [Thorup, Combinatorica'07].

cs.DS↗

arXiv2023

Faster Isomorphism for $p$-Groups of Class 2 and Exponent $p$

The group isomorphism problem determines whether two groups, given by their Cayley tables, are isomorphic. For groups with order $n$, an algorithm with $n^{(\log n + O(1))}$ running time, attributed to Tarjan, was proposed in the 1970s [Mil78]. Despite the extensive study over the past decades, the current best group isomorphism algorithm has an $n^{(1 / 4 + o(1))\log n}$ running time [Ros13]. The isomorphism testing for $p$-groups of (nilpotent) class 2 and exponent $p$ has been identified as a major barrier to obtaining an $n^{o(\log n)}$ time algorithm for the group isomorphism problem. Although the $p$-groups of class 2 and exponent $p$ have much simpler algebraic structures than general groups, the best-known isomorphism testing algorithm for this group class also has an $n^{O(\log n)}$ running time. In this paper, we present an isomorphism testing algorithm for $p$-groups of class 2 and exponent $p$ with running time $n^{O((\log n)^{5/6})}$ for any prime $p > 2$. Our result is based on a novel reduction to the skew-symmetric matrix tuple isometry problem [IQ19]. To obtain the reduction, we develop several tools for matrix space analysis, including a matrix space individualization-refinement method and a characterization of the low rank matrix spaces.

cs.DS↗

arXiv2022

Dynamic Kernel Graph Sparsifiers

A geometric graph associated with a set of points $P= \{x_1, x_2, \cdots, x_n \} \subset \mathbb{R}^d$ and a fixed kernel function $\mathsf{K}:\mathbb{R}^d\times \mathbb{R}^d\to\mathbb{R}_{\geq 0}$ is a complete graph on $P$ such that the weight of edge $(x_i, x_j)$ is $\mathsf{K}(x_i, x_j)$. We present a fully-dynamic data structure that maintains a spectral sparsifier of a geometric graph under updates that change the locations of points in $P$ one at a time. The update time of our data structure is $n^{o(1)}$ with high probability, and the initialization time is $n^{1+o(1)}$. Under certain assumption, our data structure can be made robust against adaptive adversaries, which makes our sparsifier applicable in iterative optimization algorithms. We further show that the Laplacian matrices corresponding to geometric graphs admit a randomized sketch for maintaining matrix-vector multiplication and projection in $n^{o(1)}$ time, under sparse updates to the query vectors, or under modification of points in $P$.

cs.DS↗

arXiv2022

Fully Dynamic s-t Edge Connectivity in Subpolynomial Time

We present a deterministic fully dynamic algorithm to answer $c$-edge connectivity queries on pairs of vertices in $n^{o(1)}$ worst case update and query time for any positive integer $c = (\log n)^{o(1)}$ for a graph with $n$ vertices. Previously, only polylogarithmic and $O(\sqrt{n})$ worst case update time fully dynamic algorithms were known for answering $1$, $2$ and $3$-edge connectivity queries respectively [Henzinger and King 1995, Frederikson 1997, Galil and Italiano 1991]. Our result extends the $c$-edge connectivity vertex sparsifier [Chalermsook et al. 2021] to a multi-level sparsification framework. As our main technical contribution, we present a novel update algorithm for the multi-level $c$-edge connectivity vertex sparsifier with subpolynomial update time.

cs.DS↗

arXiv2022

Minor Sparsifiers and the Distributed Laplacian Paradigm

We study distributed algorithms built around minor-based vertex sparsifiers, and give the first algorithm in the CONGEST model for solving linear systems in graph Laplacian matrices to high accuracy. Our Laplacian solver has a round complexity of $O(n^{o(1)}(\sqrt{n}+D))$, and thus almost matches the lower bound of $\widetildeΩ(\sqrt{n}+D)$, where $n$ is the number of nodes in the network and $D$ is its diameter. We show that our distributed solver yields new sublinear round algorithms for several cornerstone problems in combinatorial optimization. This is achieved by leveraging the powerful algorithmic framework of Interior Point Methods (IPMs) and the Laplacian paradigm in the context of distributed graph algorithms, which entails numerically solving optimization problems on graphs via a series of Laplacian systems. Problems that benefit from our distributed algorithmic paradigm include exact mincost flow, negative weight shortest paths, maxflow, and bipartite matching on sparse directed graphs. For the maxflow problem, this is the first exact distributed algorithm that applies to directed graphs, while the previous work by [Ghaffari et al. SICOMP'18] considered the approximate setting and works only for undirected graphs. For the mincost flow and the negative weight shortest path problems, our results constitute the first exact distributed algorithms running in a sublinear number of rounds. Given that the hybrid between IPMs and the Laplacian paradigm has proven useful for tackling numerous optimization problems in the centralized setting, we believe that our distributed solver will find future applications.

cs.DS↗

arXiv2021

Approximation Algorithms for LCS and LIS with Truly Improved Running Times

Longest common subsequence ($\mathsf{LCS}$) is a classic and central problem in combinatorial optimization. While $\mathsf{LCS}$ admits a quadratic time solution, recent evidence suggests that solving the problem may be impossible in truly subquadratic time. A special case of $\mathsf{LCS}$ wherein each character appears at most once in every string is equivalent to the longest increasing subsequence problem ($\mathsf{LIS}$) which can be solved in quasilinear time. In this work, we present novel algorithms for approximating $\mathsf{LCS}$ in truly subquadratic time and $\mathsf{LIS}$ in truly sublinear time. Our approximation factors depend on the ratio of the optimal solution size over the input size. We denote this ratio by $λ$ and obtain the following results for $\mathsf{LCS}$ and $\mathsf{LIS}$ without any prior knowledge of $λ$. $\bullet$ A truly subquadratic time algorithm for $\mathsf{LCS}$ with approximation factor $Ω(λ^3)$. $\bullet$A truly sublinear time algorithm for $\mathsf{LIS}$ with approximation factor $Ω(λ^3)$. Triangle inequality was recently used by [Boroujeni, Ehsani, Ghodsi, HajiAghayi and Seddighin SODA 2018] and [Charkraborty, Das, Goldenberg, Koucky and Saks FOCS 2018] to present new approximation algorithms for edit distance. Our techniques for $\mathsf{LCS}$ extend the notion of triangle inequality to non-metric settings.

cs.DS↗

arXiv2021

Universally-Optimal Distributed Shortest Paths and Transshipment via Graph-Based L1-Oblivious Routing

We provide universally-optimal distributed graph algorithms for $(1+\varepsilon)$-approximate shortest path problems including shortest-path-tree and transshipment. The universal optimality of our algorithms guarantees that, on any $n$-node network $G$, our algorithm completes in $T \cdot n^{o(1)}$ rounds whenever a $T$-round algorithm exists for $G$. This includes $D \cdot n^{o(1)}$-round algorithms for any planar or excluded-minor network. Our algorithms never require more than $(\sqrt{n} + D) \cdot n^{o(1)}$ rounds, resulting in the first sub-linear-round distributed algorithm for transshipment. The key technical contribution leading to these results is the first efficient $n^{o(1)}$-competitive linear $\ell_1$-oblivious routing operator that does not require the use of $\ell_1$-embeddings. Our construction is simple, solely based on low-diameter decompositions, and -- in contrast to all known constructions -- directly produces an oblivious flow instead of just an approximation of the optimal flow cost. This also has the benefit of simplifying the interaction with Sherman's multiplicative weight framework [SODA'17] in the distributed setting and its subsequent rounding procedures.

cs.DS↗

arXiv2021

String Matching with Wildcards in the Massively Parallel Computation Model

We study distributed algorithms for string matching problem in presence of wildcard characters. Given a string T (a text), we look for all occurrences of another string P (a pattern) as a substring of string T . Each wildcard character in the pattern matches a specific class of strings based on its type. String matching is one of the most fundamental problems in computer science, especially in the fields of bioinformatics and machine learning. Persistent effort has led to a variety of algorithms for the problem since 1960s. With rise of big data and the inevitable demand to solve problems on huge data sets, there have been many attempts to adapt classic algorithms into the MPC framework to obtain further efficiency. MPC is a recent framework for parallel computation of big data, which is designed to capture the MapReduce-like algorithms. In this paper, we study the string matching problem using a set of tools translated to MPC model. We consider three types of wildcards in string matching: - '?' wildcard: In this setting, the pattern is allowed to contain special '?' characters or don't cares that match any character of the text. String matching with don't cares could be solved by fast convolutions, and we give a constant round MPC algorithm for which by utilizing FFT in a constant number of MPC rounds. - '+' wildcard: '+' wildcard is a special character that allows for arbitrary repetitions of a character. When the pattern contains '+' wildcard characters, our algorithm runs in a constant number of MPC rounds by a reduction from subset matching problem. - '*' wildcard: '*' is a special character that matches with any substring of the text. When '*' is allowed in the pattern, we solve two special cases of the problem in logarithmic rounds.

cs.DC↗

arXiv2020

On the Hardness of Massively Parallel Computation

We investigate whether there are inherent limits of parallelization in the (randomized) massively parallel computation (MPC) model by comparing it with the (sequential) RAM model. As our main result, we show the existence of hard functions that are essentially not parallelizable in the MPC model. Based on the widely-used random oracle methodology in cryptography with a cryptographic hash function $h:\{0,1\}^n \rightarrow \{0,1\}^n$ computable in time $t_h$, we show that there exists a function that can be computed in time $O(T\cdot t_h)$ and space $S$ by a RAM algorithm, but any MPC algorithm with local memory size $s < S/c$ for some $c>1$ requires at least $\tildeΩ(T)$ rounds to compute the function, even in the average case, for a wide range of parameters $n \leq S \leq T \leq 2^{n^{1/4}}$. Our result is almost optimal in the sense that by taking $T$ to be much larger than $t_h$, \textit{e.g.}, $T$ to be sub-exponential in $t_h$, to compute the function, the round complexity of any MPC algorithm with small local memory size is asymptotically the same (up to a polylogarithmic factor) as the time complexity of the RAM algorithm. Our result is obtained by adapting the so-called compression argument from the data structure lower bounds and cryptography literature to the context of massively parallel computation.

cs.DS↗

arXiv2020

Fast Noise Removal for $k$-Means Clustering

This paper considers $k$-means clustering in the presence of noise. It is known that $k$-means clustering is highly sensitive to noise, and thus noise should be removed to obtain a quality solution. A popular formulation of this problem is called $k$-means clustering with outliers. The goal of $k$-means clustering with outliers is to discard up to a specified number $z$ of points as noise/outliers and then find a $k$-means solution on the remaining data. The problem has received significant attention, yet current algorithms with theoretical guarantees suffer from either high running time or inherent loss in the solution quality. The main contribution of this paper is two-fold. Firstly, we develop a simple greedy algorithm that has provably strong worst case guarantees. The greedy algorithm adds a simple preprocessing step to remove noise, which can be combined with any $k$-means clustering algorithm. This algorithm gives the first pseudo-approximation-preserving reduction from $k$-means with outliers to $k$-means without outliers. Secondly, we show how to construct a coreset of size $O(k \log n)$. When combined with our greedy algorithm, we obtain a scalable, near linear time algorithm. The theoretical contributions are verified experimentally by demonstrating that the algorithm quickly removes noise and obtains a high-quality clustering.

cs.DS↗

arXiv2020

Approximating LCS in Linear Time: Beating the $\sqrt{n}$ Barrier

Longest common subsequence (LCS) is one of the most fundamental problems in combinatorial optimization. Apart from theoretical importance, LCS has enormous applications in bioinformatics, revision control systems, and data comparison programs. Although a simple dynamic program computes LCS in quadratic time, it has been recently proven that the problem admits a conditional lower bound and may not be solved in truly subquadratic time. In addition to this, LCS is notoriously hard with respect to approximation algorithms. Apart from a trivial sampling technique that obtains a $n^{x}$ approximation solution in time $O(n^{2-2x})$ nothing else is known for LCS. This is in sharp contrast to its dual problem edit distance for which several linear time solutions are obtained in the past two decades.

cs.DS↗

arXiv2019

Parallel Batch-Dynamic Graphs: Algorithms and Lower Bounds

In this paper we study the problem of dynamically maintaining graph properties under batches of edge insertions and deletions in the massively parallel model of computation. In this setting, the graph is stored on a number of machines, each having space strongly sublinear with respect to the number of vertices, that is, $n^ε$ for some constant $0 < ε< 1$. Our goal is to handle batches of updates and queries where the data for each batch fits onto one machine in constant rounds of parallel computation, as well as to reduce the total communication between the machines. This objective corresponds to the gradual buildup of databases over time, while the goal of obtaining constant rounds of communication for problems in the static setting has been elusive for problems as simple as undirected graph connectivity. We give an algorithm for dynamic graph connectivity in this setting with constant communication rounds and communication cost almost linear in terms of the batch size. Our techniques combine a new graph contraction technique, an independent random sample extractor from correlated samples, as well as distributed data structures supporting parallel updates and queries in batches. We also illustrate the power of dynamic algorithms in the MPC model by showing that the batched version of the adaptive connectivity problem is $\mathsf{P}$-complete in the centralized setting, but sub-linear sized batches can be handled in a constant number of rounds. Due to the wide applicability of our approaches, we believe it represents a practically-motivated workaround to the current difficulties in designing more efficient massively parallel static graph algorithms.

cs.DS↗

arXiv2018

Massively Parallel Algorithms for Finding Well-Connected Components in Sparse Graphs

A fundamental question that shrouds the emergence of massively parallel computing (MPC) platforms is how can the additional power of the MPC paradigm be leveraged to achieve faster algorithms compared to classical parallel models such as PRAM? Previous research has identified the sparse graph connectivity problem as a major obstacle to such improvement: While classical logarithmic-round PRAM algorithms for finding connected components in any $n$-vertex graph have been known for more than three decades, no $o(\log{n})$-round MPC algorithms are known for this task with truly sublinear in $n$ memory per machine. This problem arises when processing massive yet sparse graphs with $O(n)$ edges, for which the interesting setting of parameters is $n^{1-Ω(1)}$ memory per machine. It is conjectured that achieving an $o(\log{n})$-round algorithm for connectivity on general sparse graphs with $n^{1-Ω(1)}$ per-machine memory may not be possible, and this conjecture also forms the basis for multiple conditional hardness results on the round complexity of other problems in the MPC model. We take an opportunistic approach towards the sparse graph connectivity problem, by designing an algorithm with improved performance guarantees in terms of the connectivity structure of the input graph. Formally, we design an algorithm that finds all connected components with spectral gap at least $λ$ in a graph in $O(\log\log{n} + \log{(1/λ)})$ MPC rounds and $n^{Ω(1)}$ memory per machine. As such, this algorithm achieves an exponential round reduction on sparse "well-connected" components (i.e., $λ\geq 1/\text{polylog}{(n)}$) using only $n^{Ω(1)}$ memory per machine and $\widetilde{O}(n)$ total memory, and still operates in $o(\log n)$ rounds even when $λ= 1/n^{o(1)}$.

cs.DS↗

arXiv2017

Linear regression without correspondence

This article considers algorithmic and statistical aspects of linear regression when the correspondence between the covariates and the responses is unknown. First, a fully polynomial-time approximation scheme is given for the natural least squares optimization problem in any constant dimension. Next, in an average-case and noise-free setting where the responses exactly correspond to a linear function of i.i.d. draws from a standard multivariate normal distribution, an efficient algorithm based on lattice basis reduction is shown to exactly recover the unknown linear function in arbitrary dimension. Finally, lower bounds on the signal-to-noise ratio are established for approximate recovery of the unknown linear function by any estimator.

cs.LG↗

arXiv2017

Local discontinuous Galerkin methods for the time tempered fractional diffusion equation

In this article, we consider discrete schemes for a fractional diffusion equation involving a tempered fractional derivative in time. We present a semi-discrete scheme by using the local discontinuous Galerkin (LDG) discretization in the spatial variables. We prove that the semi-discrete scheme is unconditionally stable in $L^2$ norm and convergence with optimal convergence rate $\mathcal{O}(h^{k+1})$. We develop a class of fully discrete LDG schemes by combining the weighted and shifted Lubich difference operators with respect to the time variable, and establish the error estimates. Finally, numerical experiments are presented to verify the theoretical results.

math.NA↗

arXiv2016

Structure and automorphisms of primitive coherent configurations

Coherent configurations (CCs) are highly regular colorings of the set of ordered pairs of a "vertex set"; each color represents a "constituent digraph." CCs arise in the study of permutation groups, combinatorial structures such as partially balanced designs, and the analysis of algorithms; their history goes back to Schur in the 1930s. A CC is primitive (PCC) if all its constituent digraphs are connected. We address the problem of classifying PCCs with large automorphism groups. This project was started in Babai's 1981 paper in which he showed that only the trivial PCC admits more than $\exp(\tilde{O}(n^{1/2}))$ automorphisms. (Here, $n$ is the number of vertices and the $\tilde{O}$ hides polylogarithmic factors.) In the present paper we classify all PCCs with more than $\exp(\tilde{O}(n^{1/3}))$ automorphisms, making the first progress on Babai's conjectured classification of all PCCs with more than $\exp(n^ε)$ automorphisms. A corollary to Babai's 1981 result solved a then 100-year-old problem on primitive but not doubly transitive permutation groups, giving an $\exp(\tilde{O}(n^{1/2}))$ bound on their order. In a similar vein, our result implies an $\exp(\tilde{O}(n^{1/3}))$ upper bound on the order of such groups, with known exceptions. This improvement of Babai's result was previously known only through the Classification of Finite Simple Groups (Cameron, 1981), while our proof, like Babai's, is elementary and almost purely combinatorial. Our analysis relies on a new combinatorial structure theory we develop for PCCs. In particular, we demonstrate the presence of "asymptotically uniform clique geometries" on PCCs in a certain range of the parameters.

math.CO↗