arXiv subjects

Adam Karczmarz

Publications and source records attributed to Adam Karczmarz.

At least 19 recordsLinked to original sources

arXiv2026

Strongly Polynomial Parallel Maximum Flow Revisited

We study the maximum flow problem in directed networks with real capacities in the parallel setting. For a network with $n$ vertices and $m$ arcs, we show that a randomized parallel implementation of a variant of the strongly polynomial max-flow algorithm of Dadush, Orlin, Sidford, and V\'egh [SODA 2026] runs in $\tilde{O}(mn)$ work and $\tilde{O}(m)$ depth. This improves upon the previously described tradeoffs between work and depth for strongly polynomial parallel maximum flow algorithms: earlier $\tilde{O}(n^3)$-work algorithms have $\tilde{O}(n^2)$ depth [Shiloach and Vishkin, J. Algorithms 1982; Goldberg and Tarjan, J. ACM 1988], while the known $\tilde{O}(m)$-depth approach uses $\tilde{O}(mn^3)$ work [Orlin, Oper. Res. 1993].

cs.DS↗

arXiv2025

Strongly Polynomial Parallel Work-Depth Tradeoffs for Directed SSSP

In this paper, we show new strongly polynomial work-depth tradeoffs for computing single-source shortest paths (SSSP) in non-negatively weighted directed graphs in parallel. Most importantly, we prove that directed SSSP can be solved within $\tilde{O}(m+n^{2-\epsilon})$ work and $\tilde{O}(n^{1-\epsilon})$ depth for some positive $\epsilon>0$. In particular, for dense graphs with non-negative real weights, we provide the first nearly work-efficient strongly polynomial algorithm with sublinear depth. Our result immediately yields improved strongly polynomial parallel algorithms for min-cost flow and the assignment problem. It also leads to the first non-trivial strongly polynomial dynamic algorithm for minimum mean cycle. Moreover, we develop efficient parallel algorithms in the Word RAM model for several variants of SSSP in graphs with exponentially large edge weights.

cs.DS↗

arXiv2025

VARSHAP: Addressing Global Dependency Problems in Explainable AI with Variance-Based Local Feature Attribution

Existing feature attribution methods like SHAP often suffer from global dependence, failing to capture true local model behavior. This paper introduces VARSHAP, a novel model-agnostic local feature attribution method which uses the reduction of prediction variance as the key importance metric of features. Building upon Shapley value framework, VARSHAP satisfies the key Shapley axioms, but, unlike SHAP, is resilient to global data distribution shifts. Experiments on synthetic and real-world datasets demonstrate that VARSHAP outperforms popular methods such as KernelSHAP or LIME, both quantitatively and qualitatively.

cs.LG↗

arXiv2025

Fully Dynamic Algorithms for Transitive Reduction

Given a directed graph $G$, a transitive reduction $G^t$ of $G$ (first studied by Aho, Garey, Ullman [SICOMP `72]) is a minimal subgraph of $G$ that preserves the reachability relation between every two vertices in $G$. In this paper, we study the computational complexity of transitive reduction in the dynamic setting. We obtain the first fully dynamic algorithms for maintaining a transitive reduction of a general directed graph undergoing updates such as edge insertions or deletions. Our first algorithm achieves $O(m+n \log n)$ amortized update time, which is near-optimal for sparse directed graphs, and can even support extended update operations such as inserting a set of edges all incident to the same vertex, or deleting an arbitrary set of edges. Our second algorithm relies on fast matrix multiplication and achieves $O(m+ n^{1.585})$ \emph{worst-case} update time.

cs.DS↗

arXiv2025

On Incremental Approximate Shortest Paths in Directed Graphs

In this paper, we show new data structures maintaining approximate shortest paths in sparse directed graphs with polynomially bounded non-negative edge weights under edge insertions. We give more efficient incremental $(1+\epsilon)$-approximate APSP data structures that work against an adaptive adversary: a deterministic one with $\tilde{O}(m^{3/2}n^{3/4})$ total update time and a randomized one with $\tilde{O}(m^{4/3}n^{5/6})$ total update time. For sparse graphs, these both improve polynomially upon the best-known bound against an adaptive adversary. To achieve that, building on the ideas of [Chechik-Zhang, SODA'21] and [Kyng-Meierhans-Probst Gutenberg, SODA'22], we show a near-optimal $(1+\epsilon)$-approximate incremental SSSP data structure for a special case when all edge updates are adjacent to the source, that might be of independent interest. We also describe a very simple and near-optimal \emph{offline} incremental $(1+\epsilon)$-approximate SSSP data structure. While online near-linear partially dynamic SSSP data structures have been elusive so far (except for dense instances), our result excludes using certain types of impossibility arguments to rule them out. Additionally, our offline solution leads to near-optimal and deterministic all-pairs bounded-leg shortest paths data structure for sparse graphs.

cs.DS↗

arXiv2024

Towards Scalable and Practical Batch-Dynamic Connectivity

We study the problem of dynamically maintaining the connected components of an undirected graph subject to edge insertions and deletions. We give the first parallel algorithm for the problem which is work-efficient, supports batches of updates, runs in polylogarithmic depth, and uses only linear total space. The existing algorithms for the problem either use super-linear space, do not come with strong theoretical bounds, or are not parallel. On the empirical side, we provide the first implementation of the cluster forest algorithm, the first linear-space and poly-logarithmic update time algorithm for dynamic connectivity. Experimentally, we find that our algorithm uses up to 19.7x less space and is up to 6.2x faster than the level-set algorithm of HDT, arguably the most widely-implemented dynamic connectivity algorithm with strong theoretical guarantees.

cs.DS↗

arXiv2024

Faster Approximation Algorithms for Restricted Shortest Paths in Directed Graphs

In the restricted shortest paths problem, we are given a graph $G$ whose edges are assigned two non-negative weights: lengths and delays, a source $s$, and a delay threshold $D$. The goal is to find, for each target $t$, the length of the shortest $(s,t)$-path whose total delay is at most $D$. While this problem is known to be NP-hard [Garey and Johnson, 1979] $(1+\varepsilon)$-approximate algorithms running in $\tilde{O}(mn)$ time [Goel et al., INFOCOM'01; Lorenz and Raz, Oper. Res. Lett.'01] given more than twenty years ago have remained the state-of-the-art for directed graphs. An open problem posed by [Bernstein, SODA'12] -- who gave a randomized $m\cdot n^{o(1)}$ time bicriteria $(1+\varepsilon, 1+\varepsilon)$-approximation algorithm for undirected graphs -- asks if there is similarly an $o(mn)$ time approximation scheme for directed graphs. We show two randomized bicriteria $(1+\varepsilon, 1+\varepsilon)$-approximation algorithms that give an affirmative answer to the problem: one suited to dense graphs, and the other that works better for sparse graphs. On directed graphs with a quasi-polynomial weights aspect ratio, our algorithms run in time $\tilde{O}(n^2)$ and $\tilde{O}(mn^{3/5})$ or better, respectively. More specifically, the algorithm for sparse digraphs runs in time $\tilde{O}(mn^{(3 - \alpha)/5})$ for graphs with $n^{1 + \alpha}$ edges for any real $\alpha \in [0,1/2]$.

cs.DS↗

arXiv2024

Subquadratic algorithms in minor-free digraphs: (weighted) distance oracles, decremental reachability, and more

Le and Wulff-Nilsen [SODA '24] initiated a systematic study of VC set systems to unweighted $K_h$-minor-free directed graphs. We extend their results in the following ways: $\bullet$ We present the first application of VC set systems for real-weighted minor-free digraphs to build the first exact subquadratic-space distance oracle with $O(\log n)$ query time. Prior work using VC set systems only applied in unweighted and integer weighted digraphs. $\bullet$ We describe a unified system for analyzing the VC dimension of balls and the LP set system (based on Li--Parter [STOC '19]) of Le--Wulff-Nilsen [SODA '24] using pseudodimension. This is a major conceptual contribution that allows for both improving our understanding of set systems in digraphs as well as improving the bound of the LP set system in directed graphs to $h-1$. $\bullet$ We present the first application of these set systems in a dynamic setting. Specifically, we construct decremental reachability oracles with subquadratic total update time and constant query time. Prior to this work, it was not known if this was possible to construct oracles with subquadratic total update time and polylogarithmic query time, even in planar digraphs. $\bullet$ We describe subquadratic time algorithms for unweighted digraphs including (1) constructions of exact distance oracles, (2) computation of vertex eccentricities and Wiener index. The main innovation in obtaining these results is the use of dynamic string data structures.

cs.DS↗

arXiv2024

Fully Dynamic Shortest Paths in Sparse Digraphs

We study the exact fully dynamic shortest paths problem. For real-weighted directed graphs, we show a deterministic fully dynamic data structure with $\tilde{O}(mn^{4/5})$ worst-case update time processing arbitrary $s,t$-distance queries in $\tilde{O}(n^{4/5})$ time. This constitutes the first non-trivial update/query tradeoff for this problem in the regime of sparse weighted directed graphs.

cs.DS↗

arXiv2024

Fully Dynamic Strongly Connected Components in Planar Digraphs

In this paper, we consider maintaining strongly connected components (SCCs) of a directed planar graph subject to edge insertions and deletions. We show a data structure maintaining an implicit representation of the SCCs within $\tilde{O}(n^{6/7})$ worst-case time per update. The data structure supports, in $O(\log^2{n})$ time, reporting vertices of any specified SCC (with constant overhead per reported vertex) and aggregating vertex information (e.g., computing the maximum label) over all the vertices of that SCC. Furthermore, it can maintain global information about the structure of SCCs, such as the number of SCCs or the size of the largest SCC. To the best of our knowledge, no fully dynamic SCCs data structures with sublinear update time have been previously known for any major subclass of digraphs. Our result should be contrasted with the known $n^{1-o(1)}$ amortized update time lower bound conditional on SETH, which holds even for dynamically maintaining whether a general digraph has more than two SCCs.

cs.DS↗

arXiv2024

Accurate estimation of feature importance faithfulness for tree models

In this paper, we consider a perturbation-based metric of predictive faithfulness of feature rankings (or attributions) that we call PGI squared. When applied to decision tree-based regression models, the metric can be computed accurately and efficiently for arbitrary independent feature perturbation distributions. In particular, the computation does not involve Monte Carlo sampling that has been typically used for computing similar metrics and which is inherently prone to inaccuracies. Moreover, we propose a method of ranking features by their importance for the tree model's predictions based on PGI squared. Our experiments indicate that in some respects, the method may identify the globally important features better than the state-of-the-art SHAP explainer

cs.LG↗

arXiv2023

Exact Shortest Paths with Rational Weights on the Word RAM

Exact computation of shortest paths in weighted graphs has been traditionally studied in one of two settings. First, one can assume that the edge weights are real numbers and all the performed operations on reals (typically comparisons and additions) take constant time. Classical Dijkstra's and Bellman-Ford algorithms have been described in this setting. More efficient exact shortest paths algorithms have been obtained for integer-weighted graphs. Integrality assumption not only enables faster algorithms but also allows implementing the aforementioned algorithms in a much more realistic word RAM model where only arithmetic operations on $O(\log{n})$-bit integers are performed in constant time. On the word RAM one can as efficiently exactly encode even \emph{rational-weighted} instances with $O(\log{n})$-bit numerators and denominators. However, the known exact real-weighted shortest paths algorithms, run on such a rational input, can easily encounter intermediate values of $\Theta(n)$ bits if represented exactly. This leads to a factor-$\Omega(n)$ slowdown on the word RAM. At the same time, the scaling algorithms suited for integer weights do not produce exact solutions for rational inputs without dramatically increasing their accuracy. In this paper, we design randomized exact single-source shortest paths algorithms for rational-weighted graphs on the word RAM. Most importantly, in the non-negative case, we obtain a near-linear time algorithm matching Dijkstra's algorithm running time up to polylogarithmic factors. In presence of negative weights, we give an $\tilde{O}(n^{2.5})$-time algorithm breaking through the best known strongly polynomial bound attained by Bellman-Ford for sufficiently dense graphs.

cs.DS↗

arXiv2023

Max $s,t$-Flow Oracles and Negative Cycle Detection in Planar Digraphs

We study the maximum $s,t$-flow oracle problem on planar directed graphs where the goal is to design a data structure answering max $s,t$-flow value (or equivalently, min $s,t$-cut value) queries for arbitrary source-target pairs $(s,t)$. For the case of polynomially bounded integer edge capacities, we describe an exact max $s,t$-flow oracle with truly subquadratic space and preprocessing, and sublinear query time. Moreover, if $(1-\epsilon)$-approximate answers are acceptable, we obtain a static oracle with near-linear preprocessing and $\tilde{O}(n^{3/4})$ query time and a dynamic oracle supporting edge capacity updates and queries in $\tilde{O}(n^{6/7})$ worst-case time. To the best of our knowledge, for directed planar graphs, no (approximate) max $s,t$-flow oracles have been described even in the unweighted case, and only trivial tradeoffs involving either no preprocessing or precomputing all the $n^2$ possible answers have been known. One key technical tool we develop on the way is a sublinear (in the number of edges) algorithm for finding a negative cycle in so-called dense distance graphs. By plugging it in earlier frameworks, we obtain improved bounds for other fundamental problems on planar digraphs. In particular, we show: (1) a deterministic $O(n\log(nC))$ time algorithm for negatively-weighted SSSP in planar digraphs with integer edge weights at least $-C$. This improves upon the previously known bounds in the important case of weights polynomial in $n$, and (2) an improved $O(n\log{n})$ bound on finding a perfect matching in a bipartite planar graph.

cs.DS↗

arXiv2023

Deterministic Fully Dynamic SSSP and More

We present the first non-trivial fully dynamic algorithm maintaining exact single-source distances in unweighted graphs. This resolves an open problem stated by Sankowski [COCOON 2005] and van den Brand and Nanongkai [FOCS 2019]. Previous fully dynamic single-source distances data structures were all approximate, but so far, non-trivial dynamic algorithms for the exact setting could only be ruled out for polynomially weighted graphs (Abboud and Vassilevska Williams, [FOCS 2014]). The exact unweighted case remained the main case for which neither a subquadratic dynamic algorithm nor a quadratic lower bound was known. Our dynamic algorithm works on directed graphs, is deterministic, and can report a single-source shortest paths tree in subquadratic time as well. Thus we also obtain the first deterministic fully dynamic data structure for reachability (transitive closure) with subquadratic update and query time. This answers an open problem of van den Brand, Nanongkai, and Saranurak [FOCS 2019]. Finally, using the same framework we obtain the first fully dynamic data structure maintaining all-pairs $(1+\epsilon)$-approximate distances within non-trivial sub-$n^\omega$ worst-case update time while supporting optimal-time approximate shortest path reporting at the same time. This data structure is also deterministic and therefore implies the first known non-trivial deterministic worst-case bound for recomputing the transitive closure of a digraph.

cs.DS↗

arXiv2023

Sensitivity and Dynamic Distance Oracles via Generic Matrices and Frobenius Form

Algebraic techniques have had an important impact on graph algorithms so far. Porting them, e.g., the matrix inverse, into the dynamic regime improved best-known bounds for various dynamic graph problems. In this paper, we develop new algorithms for another cornerstone algebraic primitive, the Frobenius normal form (FNF). We apply our developments to dynamic and fault-tolerant exact distance oracle problems on directed graphs. For generic matrices $A$ over a finite field accompanied by an FNF, we show (1) an efficient data structure for querying submatrices of the first $k\geq 1$ powers of $A$, and (2) a near-optimal algorithm updating the FNF explicitly under rank-1 updates. By representing an unweighted digraph using a generic matrix over a sufficiently large field (obtained by random sampling) and leveraging the developed FNF toolbox, we obtain: (a) a conditionally optimal distance sensitivity oracle (DSO) in the case of single-edge or single-vertex failures, providing a partial answer to the open question of Gu and Ren [ICALP'21], (b) a multiple-failures DSO improving upon the state of the art (vd. Brand and Saranurak [FOCS'19]) wrt. both preprocessing and query time, (c) improved dynamic distance oracles in the case of single-edge updates, and (d) a dynamic distance oracle supporting vertex updates, i.e., changing all edges incident to a single vertex, in $\tilde{O}(n^2)$ worst-case time and distance queries in $\tilde{O}(n)$ time.

cs.DS↗

arXiv2022

Subquadratic Dynamic Path Reporting in Directed Graphs Against an Adaptive Adversary

We study reachability and shortest paths problems in dynamic directed graphs. Whereas algebraic dynamic data structures supporting edge updates and reachability/distance queries have been known for quite a long time, they do not, in general, allow reporting the underlying paths within the same time bounds, especially against an adaptive adversary. In this paper we develop the first known fully dynamic reachability data structures working against an adaptive adversary and supporting edge updates and path queries for two natural variants: (1) point-to-point path reporting, and (2) single-source reachability tree reporting. For point-to-point queries in DAGs, we achieve $O(n^{1.529})$ worst-case update and query bounds, whereas for tree reporting in DAGs, the worst-case bounds are $O(n^{1.765})$. More importantly, we show how to lift these algorithms to work on general graphs at the cost of increasing the bounds to $n^{1+5/6+o(1)}$ and making the update times amortized. On the way to accomplishing that, we obtain two interesting subresults. We give subquadratic fully dynamic algorithms for topological order (in a DAG), and strongly connected components. To the best of our knowledge, such algorithms have not been described before. Additionally, we provide deterministic incremental data structures for reachability and shortest paths that handle edge insertions and report the respective paths within subquadratic worst-case time bounds. For reachability and $(1+\epsilon)$-approximate shortest paths in weighted digraphs, these bounds match the best known dynamic matrix inverse-based randomized bounds for fully dynamic reachability [v.d.Brand, Nanongkai and Saranurak, FOCS'19]. For exact shortest paths in unweighted graphs, the obtained bounds in the incremental setting polynomially improve upon the respective best known randomized update/distance query bounds in the fully dynamic setting.

cs.DS↗

arXiv2021

Optimal Decremental Connectivity in Non-Sparse Graphs

We present a dynamic algorithm for maintaining the connected and 2-edge-connected components in an undirected graph subject to edge deletions. The algorithm is Monte-Carlo randomized and processes any sequence of edge deletions in $O(m + n \operatorname{polylog} n)$ total time. Interspersed with the deletions, it can answer queries to whether any two given vertices currently belong to the same (2-edge-)connected component in constant time. Our result is based on a general Monte-Carlo randomized reduction from decremental $c$-edge-connectivity to a variant of fully-dynamic $c$-edge-connectivity on a sparse graph. While being Monte-Carlo, our reduction supports a certain final self-check that can be used in Las Vegas algorithms for static problems such as Unique Perfect Matching. For non-sparse graphs with $\Omega(n \operatorname{polylog} n)$ edges, our connectivity and $2$-edge-connectivity algorithms handle all deletions in optimal linear total time, using existing algorithms for the respective fully-dynamic problems. This improves upon an $O(m \log (n^2 / m) + n \operatorname{polylog} n)$-time algorithm of Thorup [J.Alg. 1999], which runs in linear time only for graphs with $\Omega(n^2)$ edges. Our constant amortized cost for edge deletions in decremental connectivity in non-sparse graphs should be contrasted with an $\Omega(\log n/\log\log n)$ worst-case time lower bound in the decremental setting [Alstrup, Thore Husfeldt, FOCS'98] as well as an $\Omega(\log n)$ amortized time lower-bound in the fully-dynamic setting [Patrascu and Demaine STOC'04].

cs.DS↗

arXiv2021

Improved Strongly Polynomial Algorithms for Deterministic MDPs, 2VPI Feasibility, and Discounted All-Pairs Shortest Paths

We revisit the problem of finding optimal strategies for deterministic Markov Decision Processes (DMDPs), and a closely related problem of testing feasibility of systems of $m$ linear inequalities on $n$ real variables with at most two variables per inequality (2VPI). We give a randomized trade-off algorithm solving both problems and running in $\tilde{O}(nmh+(n/h)^3)$ time using $\tilde{O}(n^2/h+m)$ space for any parameter $h\in [1,n]$. In particular, using subquadratic space we get $\tilde{O}(nm+n^{3/2}m^{3/4})$ running time, which improves by a polynomial factor upon all the known upper bounds for non-dense instances with $m=O(n^{2-\epsilon})$. Moreover, using linear space we match the randomized $\tilde{O}(nm+n^3)$ time bound of Cohen and Megiddo [SICOMP'94] that required $\tilde{\Theta}(n^2+m)$ space. Additionally, we show a new algorithm for the Discounted All-Pairs Shortest Paths problem, introduced by Madani et al. [TALG'10], that extends the DMDPs with optional end vertices. For the case of uniform discount factors, we give a deterministic algorithm running in $\tilde{O}(n^{3/2}m^{3/4})$ time, which improves significantly upon the randomized bound $\tilde{O}(n^2\sqrt{m})$ of Madani et al.

cs.DS↗