Search

In this paper, we study the Dynamic Parameterized Subset Sampling (DPSS) problem in the Word RAM model. In DPSS, the input is a set,~$S$, of~$n$ items, where each item,~$x$, has a non-negative integer weight,~$w(x)$. Given a pair of query parameters, $(\alpha, \beta)$, each of which is a non-negative rational number, a parameterized subset sampling query on~$S$ seeks to return a subset $T \subseteq S$ such that each item $x \in S$ is selected in~$T$, independently, with probability $p_x(\alpha, \beta) = \min \left\{\frac{w(x)}{\alpha \sum_{x\in S} w(x)+\beta}, 1 \right\}$. More specifically, the DPSS problem is defined in a dynamic setting, where the item set,~$S$, can be updated with insertions of new items or deletions of existing items. Our first main result is an optimal algorithm for solving the DPSS problem, which achieves~$O(n)$ pre-processing time, $O(1+\mu_S(\alpha,\beta))$ expected time for each query parameterized by $(\alpha, \beta)$, given on-the-fly, and $O(1)$ time for each update; here, $\mu_S(\alpha,\beta)$ is the expected size of the query result. At all times, the worst-case space consumption of our algorithm is linear in the current number of items in~$S$. Our second main contribution is a hardness result for the DPSS problem when the item weights are~$O(1)$-word float numbers, rather than integers. Specifically, we reduce Integer Sorting to the deletion-only DPSS problem with float item weights. Our reduction implies that an optimal algorithm for deletion-only DPSS with float item weights (achieving all the same bounds as aforementioned) implies an optimal algorithm for Integer Sorting. The latter remains an important open problem. Last but not least, a key technical ingredient for our first main result is an efficient algorithm for generating Truncated Geometric random variates in $O(1)$ expected time in the Word RAM model., Comment: 29 pages, 10 figures, to be published in PODS25

We consider the Max Unique Coverage problem, including applications to the data stream model. The input is a universe of $n$ elements, a collection of $m$ subsets of this universe, and a cardinality constraint, $k$. The goal is to select a subcollection of at most $k$ sets that maximizes unique coverage, i.e, the number of elements contained in exactly one of the selected sets. The Max Unique Coverage problem has applications in wireless networks, radio broadcast, and envy-free pricing. Our first main result is a fixed-parameter tractable approximation scheme (FPT-AS) for Max Unique Coverage, parameterized by $k$ and the maximum element frequency, $r$, which can be implemented on a data stream. Our FPT-AS finds a $(1-\epsilon)$-approximation while maintaining a kernel of size $\tilde{O}(k r/\epsilon)$, which can be combined with subsampling to use $\tilde{O}(k^2 r / \epsilon^3)$ space overall. This significantly improves on the previous-best FPT-AS with the same approximation, but a kernel of size $\tilde{O}(k^2 r / \epsilon^2)$. In order to achieve our result, we show upper bounds on the ratio of a collection's coverage to the unique coverage of a maximizing subcollection; this is by constructing explicit algorithms that find a subcollection with unique coverage at least a logarithmic ratio of the collection's coverage. We complement our algorithms with our second main result, showing that $\Omega(m / k^2)$ space is necessary to achieve a $(1.5 + o(1))/(\ln k - 1)$-approximation in the data stream. This dramatically improves the previous-best lower bound showing that $\Omega(m / k^2)$ is necessary to achieve better than a $e^{-1+1/k}$-approximation., Comment: 26 pages. Accepted into APPROX 2024

Maximizing a non-negative, monontone, submodular function $f$ over $n$ elements under a cardinality constraint $k$ (SMCC) is a well-studied NP-hard problem. It has important applications in, e.g., machine learning and influence maximization. Though the theoretical problem admits polynomial-time approximation algorithms, solving it in practice often involves frequently querying submodular functions that are expensive to compute. This has motivated significant research into designing parallel approximation algorithms in the adaptive complexity model; adaptive complexity (adaptivity) measures the number of sequential rounds of $\text{poly}(n)$ function queries an algorithm requires. The state-of-the-art algorithms can achieve $(1-\frac{1}{e}-\varepsilon)$-approximate solutions with $O(\frac{1}{\varepsilon^2}\log n)$ adaptivity, which approaches the known adaptivity lower-bounds. However, the $O(\frac{1}{\varepsilon^2} \log n)$ adaptivity only applies to maximizing worst-case functions that are unlikely to appear in practice. Thus, in this paper, we consider the special class of $p$-superseparable submodular functions, which places a reasonable constraint on $f$, based on the parameter $p$, and is more amenable to maximization, while also having real-world applicability. Our main contribution is the algorithm LS+GS, a finer-grained version of the existing LS+PGB algorithm, designed for instances of SMCC when $f$ is $p$-superseparable; it achieves an expected $(1-\frac{1}{e}-\varepsilon)$-approximate solution with $O(\frac{1}{\varepsilon^2}\log(p k))$ adaptivity independent of $n$. Additionally, unrelated to $p$-superseparability, our LS+GS algorithm uses only $O(\frac{n}{\varepsilon} + \frac{\log n}{\varepsilon^2})$ oracle queries, which has an improved dependence on $\varepsilon^{-1}$ over the state-of-the-art LS+PGB; this is achieved through the design of a novel thresholding subroutine., Comment: 36 pages. To be published in Approximation and Online Algorithms (Proceedings of the 21st International Workshop, WAOA 2023)

Detecting beneficial feature interactions is essential in recommender systems, and existing approaches achieve this by examining all the possible feature interactions. However, the cost of examining all the possible higher-order feature interactions is prohibitive (exponentially growing with the order increasing). Hence existing approaches only detect limited order (e.g., combinations of up to four features) beneficial feature interactions, which may miss beneficial feature interactions with orders higher than the limitation. In this paper, we propose a hypergraph neural network based model named HIRS. HIRS is the first work that directly generates beneficial feature interactions of arbitrary orders and makes recommendation predictions accordingly. The number of generated feature interactions can be specified to be much smaller than the number of all the possible interactions and hence, our model admits a much lower running time. To achieve an effective algorithm, we exploit three properties of beneficial feature interactions, and propose deep-infomax-based methods to guide the interaction generation. Our experimental results show that HIRS outperforms state-of-the-art algorithms by up to 5% in terms of recommendation accuracy., Comment: KDD 2022, 11 pages, 12 figures, 5 tables

Personalized PageRank (PPR) is a popular node proximity metric in graph mining and network research. Given a graph G=(V,E) and a source node $s \in V$, a single-source PPR (SSPPR) query asks for the PPR value $\vpi(u)$ with respect to s, which represents the relative importance of node u in the context of the source node s. Among existing algorithms for SSPPR queries, LocalPush is a fundamental method which serves as a cornerstone for subsequent algorithms. In LocalPush, a push operation is a crucial primitive operation, which distributes the probability at a node u to ALL u's neighbors via the corresponding edges. Although this push operation works well on unweighted graphs, unfortunately, it can be rather inefficient on weighted graphs. In particular, on unbalanced weighted graphs where only a few of these edges take the majority of the total weight among them, the push operation would have to distribute insignificant probabilities along those edges which just take the minor weights, resulting in expensive overhead. To resolve this issue, we propose the EdgePush algorithm, a novel method for computing SSPPR queries on weighted graphs. EdgePush decomposes the aforementioned push operations in edge-based push, allowing the algorithm to operate at the edge level granularity. Hence, it can flexibly distribute the probabilities according to edge weights. Furthermore, our EdgePush allows a fine-grained termination threshold for each individual edge, leading to a superior complexity over LocalPush. Notably, we prove that EdgePush improves the theoretical query cost of LocalPush by an order of up to O(n) when the graph's weights are unbalanced, both in terms of $\ell_1$-error and normalized additive error. Our experimental results demonstrate that EdgePush significantly outperforms state-of-the-art baselines in terms of query efficiency on large motif-based and real-world weighted graphs., Comment: VLDB 2022, volume 15, issue 7

Structural Clustering ($DynClu$) is one of the most popular graph clustering paradigms. In this paper, we consider $StrClu$ under two commonly adapted similarities, namely Jaccard similarity and cosine similarity on a dynamic graph, $G = \langle V, E\rangle$, subject to edge insertions and deletions (updates). The goal is to maintain certain information under updates, so that the $StrClu$ clustering result on~$G$ can be retrieved in $O(|V| + |E|)$ time, upon request. The state-of-the-art worst-case cost is $O(|V|)$ per update; we improve this update-time bound significantly with the $\rho$-approximate notion. Specifically, for a specified failure probability, $\delta^*$, and every sequence of $M$ updates (no need to know $M$'s value in advance), our algorithm, $DynELM$, achieves $O(\log^2 |V| + \log |V| \cdot \log \frac{M}{\delta^*})$ amortized cost for each update, at all times in linear space. Moreover, $DynELM$ provides a provable "sandwich" guarantee on the clustering quality at all times after \emph{each update} with probability at least $1 - \delta^*$. We further develop $DynELM$ into our ultimate algorithm, $DynStrClu$, which also supports cluster-group-by queries. Given $Q\subseteq V$, this puts the non-empty intersection of $Q$ and each $StrClu$ cluster into a distinct group. $DynStrClu$ not only achieves all the guarantees of $DynELM$, but also runs cluster-group-by queries in $O(|Q|\cdot \log |V|)$ time. We demonstrate the performance of our algorithms via extensive experiments, on 15 real datasets. Experimental results confirm that our algorithms are up to three orders of magnitude more efficient than state-of-the-art competitors, and still provide quality structural clustering results. Furthermore, we study the difference between the two similarities w.r.t. the quality of approximate clustering results.

User and item attributes are essential side-information; their interactions (i.e., their co-occurrence in the sample data) can significantly enhance prediction accuracy in various recommender systems. We identify two different types of attribute interactions, inner interactions and cross interactions: inner interactions are those between only user attributes or those between only item attributes; cross interactions are those between user attributes and item attributes. Existing models do not distinguish these two types of attribute interactions, which may not be the most effective way to exploit the information carried by the interactions. To address this drawback, we propose a neural Graph Matching based Collaborative Filtering model (GMCF), which effectively captures the two types of attribute interactions through modeling and aggregating attribute interactions in a graph matching structure for recommendation. In our model, the two essential recommendation procedures, characteristic learning and preference matching, are explicitly conducted through graph learning (based on inner interactions) and node matching (based on cross interactions), respectively. Experimental results show that our model outperforms state-of-the-art models. Further studies verify the effectiveness of GMCF in improving the accuracy of recommendation., Comment: 10 pages, 6 figures, 4 tables, SIGIR 2021

Personalized PageRank (PPR) is a critical measure of the importance of a node t to a source node s in a graph. The Single-Source PPR (SSPPR) query computes the PPR's of all the nodes with respect to s on a directed graph $G$ with $n$ nodes and $m$ edges, and it is an essential operation widely used in graph applications. In this paper, we propose novel algorithms for solving two variants of SSPPR: (i) high-precision queries and (ii) approximate queries. For high-precision queries, Power Iteration (PowItr) and Forward Push (FwdPush) are two fundamental approaches. Given an absolute error threshold $\lambda$, the only known bound of FwdPush is $O(\frac{m}{\lambda})$, much worse than the $O(m \log \frac{1}{\lambda})$-bound of PowItr. Whether FwdPush can achieve the same running time bound as PowItr does still remains an open question in the research community. We give a positive answer to this question by showing that the running time of a common implementation of FwdPush is actually bounded by $O(m \cdot \log \frac{1}{\lambda})$.Based on this finding, we propose a new algorithm, called Power Iteration with Forward Push (PowerPush), which incorporates the strengths of both PowItr and FwdPush. For approximate queries (with a relative error $\epsilon$), we propose a new algorithm, called SpeedPPR, with overall expected time bounded by $O(n \cdot \log n \cdot \log \frac{1}{\epsilon})$ on scale-free graphs. This bound greatly improves the $O(\frac{n \cdot \log n}{\epsilon})$ bound of a state-of-the-art algorithm FORA., Comment: 12 pages

Inspired by the great success of machine learning in the past decade, people have been thinking about the possibility of improving the theoretical results by exploring data distribution. In this paper, we revisit a fundamental problem called Distributed Tracking (DT) under an assumption that the data follows a certain (known or unknown) distribution, and propose a number data-dependent algorithms with improved theoretical bounds. Informally, in the DT problem, there is a coordinator and k players, where the coordinator holds a threshold N and each player has a counter. At each time stamp, at most one counter can be increased by one. The job of the coordinator is to capture the exact moment when the sum of all these k counters reaches N. The goal is to minimise the communication cost. While our first type of algorithms assume the concrete data distribution is known in advance, our second type of algorithms can learn the distribution on the fly. Both of the algorithms achieve a communication cost bounded byO(k log log N) with high probability, improving the state-of-the-art data-independent bound O(k log N/k). We further propose a number of implementation optimisation heuristics to improve both efficiency and robustness of the algorithms. Finally, we conduct extensive experiments on three real datasets and four synthetic datasets. The experimental results show that the communication cost of our algorithms is as least as 20% of that of the state-of-the-art algorithms.

Personalized PageRank (PPR) is a widely used node proximity measure in graph mining and network analysis. Given a source node $s$ and a target node $t$, the PPR value $\pi(s,t)$ represents the probability that a random walk from $s$ terminates at $t$, and thus indicates the bidirectional importance between $s$ and $t$. The majority of the existing work focuses on the single-source queries, which asks for the PPR value of a given source node $s$ and every node $t \in V$. However, the single-source query only reflects the importance of each node $t$ with respect to $s$. In this paper, we consider the {\em single-target PPR query}, which measures the opposite direction of importance for PPR. Given a target node $t$, the single-target PPR query asks for the PPR value of every node $s\in V$ to a given target node $t$. We propose RBS, a novel algorithm that answers approximate single-target queries with optimal computational complexity. We show that RBS improves three concrete applications: heavy hitters PPR query, single-source SimRank computation, and scalable graph neural networks. We conduct experiments to demonstrate that RBS outperforms the state-of-the-art algorithms in terms of both efficiency and precision on real-world benchmark datasets., Comment: ACM SIGKDD 2020

Resolution parameters in graph clustering represent a size and quality trade-off. We address the task of efficiently solving a parameterized graph clustering objective for all values of a resolution parameter. Specifically, we consider an objective we call LambdaPrime, involving a parameter $\lambda \in (0,1)$. This objective is related to other parameterized clustering problems, such as parametric generalizations of modularity, and captures a number of specific clustering problems as special cases, including sparsest cut and cluster deletion. While previous work provides approximation results for a single resolution parameter, we seek a set of approximately optimal clusterings for all values of $\lambda$ in polynomial time. In particular, we ask the question, how small a family of clusterings suffices to optimize -- or to approximately optimize -- the LambdaPrime objective over the full possible spectrum of $\lambda$? We obtain a family of logarithmically many clusterings by solving the parametric linear programming relaxation of LambdaPrime at a logarithmic number of parameter values, and round their solutions using existing approximation algorithms. We prove that this number is tight up to a constant factor. Specifically, for a certain class of ring graphs, a logarithmic number of feasible solutions is required to provide a constant-factor approximation for the LambdaPrime LP relaxation in all parameter regimes. We additionally show that for any graph with $n$ nodes and $m$ edges, there exists a set of $m$ or fewer clusterings such that for every $\lambda \in (0,1)$, the family contains an exact solution to the LambdaPrime objective. There also exists a set of $O(\log n)$ clusterings that provide a $(1+\varepsilon)$-approximate solution in all parameter regimes; we demonstrate simple graph classes for which these bounds are tight.