arXiv subjects

Jannik Matuschke

Publications and source records attributed to Jannik Matuschke.

At least 19 recordsLinked to original sources

arXiv2026

Robust Deterministic Policies for Markov Decision Processes under Budgeted Uncertainty

This paper studies the computation of robust deterministic policies for Markov Decision Processes (MDPs) in the Lightning Does Not Strike Twice (LDST) model of Mannor, Mebel and Xu (ICML '12). In this model, designed to provide robustness in the face of uncertain input data while not being overly conservative, transition probabilities and rewards are uncertain and the uncertainty set is constrained by a budget that limits the number of states whose parameters can deviate from their nominal values. Mannor et al. (ICML '12) showed that optimal randomized policies for MDPs in the LDST regime can be efficiently computed when only the rewards are affected by uncertainty. In contrast to these findings, we observe that the computation of optimal deterministic policies is $N\!P$-hard even when only a single terminal reward may deviate from its nominal value and the MDP consists of $2$ time periods. For this hard special case, we then derive a constant-factor approximation algorithm by combining two relaxations based on the Knapsack Cover and Generalized Assignment problem, respectively. For the general problem with possibly a large number of deviations and a longer time horizon, we derive strong inapproximability results for computing robust deterministic policies as well as $Σ_2^p$-hardness, indicating that the general problem does not even admit a compact mixed integer programming formulation.

math.OC↗

arXiv2026

The Project Scheduling Interdiction Problem with Delay Groups

Large-scale projects are frequently delayed by correlated disruptions: when a shared input such as a common supplier, a specialized team, or a supporting platform degrades, all dependent activities are slowed down simultaneously. This paper introduces delay groups to capture such disruptions: a delay group is a set of activities whose delays stem from a common cause, described jointly by an uncertainty set. Our model takes the perspective of an interdictor that, subject to a budget of $k$ groups, selects which groups to disrupt so as to maximize the project makespan. The interdictor can extend activity durations within each disrupted group by delays from a group-specific uncertainty set, while non-disrupted activities keep their nominal duration. We study the complexity of the resulting Project Scheduling Interdiction Problem with Delay Groups (PSIP-DG), which provides a worst-case stress test of the schedule. The problem is computationally intractable ($N\!P$-hard) for general polyhedral uncertainty sets, even for a single delay group with a continuous knapsack constraint. For budgeted uncertainty sets, we prove $N\!P$-hardness both when all activities of a disrupted group are delayed and when only one activity per group may be delayed, and derive an inapproximability bound of $1-1/e+ε$ for the former case. We further develop a greedy heuristic with approximation guarantee $k$ and two structure-based heuristics with initializations and neighborhoods from tractable special cases. Experiments on $5{,}000$-activity networks show that the heuristics match the solution quality of an exact solver at substantially lower running times, in some cases finding strictly better solutions.

cs.DM↗

arXiv2026

Maximum Covering Network Design on Graphs with Low Connectivity: Dynamic Programming and Block-Cut Trees

Planning accessible public services such as health care, emergency response, and schools often requires not only choosing where to open facilities but also improving the network that connects people to them, for example upgrading flood-prone roads in vulnerable regions. Most location models, however, take the network as fixed and the budget as given. We study the Maximum Covering Network Design Problem, in which a single budget is shared between opening facilities and upgrading weak links to maximize the population within a target travel distance of an open facility. The problem is hard even on the simplest networks, and planners usually want to see how coverage grows with the budget, not a single plan. We develop an exact dynamic-programming framework that exploits a property common to real road networks: their low connectivity, with many cut points whose removal disconnects the network. On trees, the recursion is self-contained: each state reduces to a few simple facility and upgrade choices that are fast to compute without a solver, giving predictable running times; for larger budgets and travel distances it outperforms solving the MILP formulation directly. On general low-connectivity networks, the framework decomposes the problem at the cut points and embeds a given MILP formulation to solve the resulting pieces, coordinating them through coverage conditions at the interfaces. This lets us compare a formulation on its own against the same formulation inside the framework: across 306 test cases the framework matches or outperforms direct solving on more than 80% of instances. Because it evaluates all budget levels in a single run, it also yields the full coverage-versus-budget curve at no extra cost, whereas direct solving must split its time across individual budgets.

math.OC↗

arXiv2026

Condorcet Dimension and Pareto Optimality for Matchings and Beyond

We study matching problems in which agents form one side of a bipartite graph and have preferences over objects on the other side. A central solution concept in this setting is popularity: a matching is popular if it is a (weak) Condorcet winner, meaning that no other matching is preferred by a strict majority of agents. It is well known, however, that Condorcet winners need not exist. We therefore turn to a natural and prominent relaxation. A set of matchings is a Condorcet-winning set if, for every competing matching, a majority of agents prefers their favorite matching in the set over the competitor. The Condorcet dimension is the smallest cardinality of a Condorcet-winning set. Our main results reveal a connection between Condorcet-winning sets and Pareto optimality. We show that any Pareto-optimal set of two matchings is, in particular, a Condorcet-winning set. This implication continues to hold when we impose matroid constraints on the set of matched objects, and even when agents' valuations are given as partial orders. The existence picture, however, changes sharply with partial orders. While for weak orders a Pareto-optimal set of two matchings always exists, this is -- surprisingly -- not the case under partial orders. Consequently, although the Condorcet dimension for matchings is 2 under weak orders (even under matroid constraints), this guarantee fails for partial orders: we prove that the Condorcet dimension is $Θ(\sqrt{n})$, and rises further to $Θ(n)$ when matroid constraints are added. On the computational side, we show that, under partial orders, deciding whether there exists a Condorcet -- winning set of a given fixed size is NP-hard. The same holds for deciding the existence of a Pareto-optimal matching, which we believe to be of independent interest. Finally, we also show that the Condorcet dimension for a related problem on arborescences is also 2.

cs.GT↗

arXiv2026

When to Identify Is to Control: On the Controllability of Combinatorial Optimization Problems

Consider a finite ground set $E$, a set of feasible solutions $X \subseteq \mathbb{R}^{E}$, and a class of objective functions $\mathcal{C}$ defined on $X$. We are interested in subsets $S$ of $E$ that control $X$ in the sense that we can induce any given solution $x \in X$ as an optimum for any given objective function $c \in \mathcal{C}$ by adding linear terms to $c$ on the coordinates corresponding to $S$. This problem has many applications, e.g., when $X$ corresponds to the set of all traffic flows, the ability to control implies that one is able to induce all target flows by imposing tolls on the edges in $S$. Our first result shows the equivalence between controllability and identifiability. If $X$ is convex, or if $X$ consists of binary vectors, then $S$ controls $X$ if and only if the restriction of $x$ to $S$ uniquely determines $x$ among all solutions in $X$. In the convex case, we further prove that the family of controlling sets forms a matroid. This structural insight yields an efficient algorithm for computing minimum-weight controlling sets from a description of the affine hull of $X$. While the equivalence extends to matroid base families, the picture changes sharply for other discrete domains. We show that when $X$ is equal to the set of $s$-$t$-paths in a directed graph, deciding whether an identifying set of a given cardinality exists is $Σ\mathsf{_2^P}$-complete. The problem remains $\mathsf{NP}$-hard even on acyclic graphs. For acyclic instances, however, we obtain an approximation guarantee by proving a tight bound on the gap between the smallest identifying sets for $X$ and its convex hull, where the latter corresponds to the $s$-$t$-flow polyhedron.

cs.DS↗

arXiv2025

Stronger Hardness for Maximum Robust Flow and Randomized Network Interdiction

We study the following fundamental network optimization problem known as Maximum Robust Flow (MRF): A planner determines a flow on $s$-$t$-paths in a given capacitated network. Then, an adversary removes $k$ arcs from the network, interrupting all flow on paths containing a removed arc. The planner's goal is to maximize the value of the surviving flow, anticipating the adversary's response (i.e., a worst-case failure of $k$ arcs). It has long been known that MRF can be solved in polynomial time when $k = 1$ (Aneja et al., 2001), whereas it is $N\!P$-hard when $k$ is part of the input (Disser and Matuschke, 2020). However, the complexity of the problem for constant values of $k > 1$ has remained elusive, in part due to structure of the natural LP description preventing the use of the equivalence of optimization and separation. This paper introduces a reduction showing that the basic version of MRF described above encapsulates the seemingly much more general variant where the adversary's choices are constrained to $k$-cliques in a compatibility graph on the arcs of the network. As a consequence of this reduction, we are able to prove the following results: (1) MRF is $N\!P$-hard for any constant number $k > 1$ of failing arcs. (2) When $k$ is part of the input, MRF is $P^{N\!P[\log]}$-hard. (3) The integer version of MRF is $Σ_2^P$-hard.

cs.DM↗

arXiv2025

Simultaneous Network Design with Restricted Link Usage

Given a digraph with two terminal vertices $s$ and $t$ as well as a conservative cost function and several not necessarily disjoint color classes on its arc set, our goal is to find a minimum-cost subset of the arcs such that its intersection with each color class contains an $s$-$t$ dipath. Problems of this type arise naturally in multi-commodity network design settings where each commodity is restricted to use links of its own color only. We study several variants of the problem, deriving strong hardness results even for restricted cases, but we also identify cases that can be solved in polynomial time. The latter ones include the cases where the color classes form a laminar family, or where the underlying digraph is acyclic and the number of color classes is constant. We also present an FPT algorithm for the general case parameterized by the number of multi-colored arcs.

cs.DS↗

arXiv2025

On the Approximability of Train Routing and the Min-Max Disjoint Paths Problem

In train routing, the headway is the minimum distance that must be maintained between successive trains for safety and robustness. We introduce a model for train routing that requires a fixed headway to be maintained between trains, and study the problem of minimizing the makespan, i.e., the arrival time of the last train, in a single-source single-sink network. For this problem, we first show that there exists an optimal solution where trains move in convoys, that is, the optimal paths for any two trains are either the same or are arc-disjoint. Via this insight, we are able to reduce the approximability of our train routing problem to that of the min-max disjoint paths problem, which asks for a collection of disjoint paths where the maximum length of any path in the collection is as small as possible. While min-max disjoint paths inherits a strong inapproximability result on directed acyclic graphs from the multi-level bottleneck assignment problem, we show that a natural greedy composition approach yields a logarithmic approximation in the number of disjoint paths for series-parallel graphs. We also present an alternative analysis of this approach that yields a guarantee depending on how often the decomposition tree of the series-parallel graph alternates between series and parallel compositions on any root-leaf path.

cs.DS↗

arXiv2023

Decomposing Probability Marginals Beyond Affine Requirements

Consider the triplet $(E, \mathcal{P}, π)$, where $E$ is a finite ground set, $\mathcal{P} \subseteq 2^E$ is a collection of subsets of $E$ and $π: \mathcal{P} \rightarrow [0,1]$ is a requirement function. Given a vector of marginals $ρ\in [0, 1]^E$, our goal is to find a distribution for a random subset $S \subseteq E$ such that $\operatorname{Pr}[e \in S] = ρ_e$ for all $e \in E$ and $\operatorname{Pr}[P \cap S \neq \emptyset] \geq π_P$ for all $P \in \mathcal{P}$, or to determine that no such distribution exists. Generalizing results of Dahan, Amin, and Jaillet, we devise a generic decomposition algorithm that solves the above problem when provided with a suitable sequence of admissible support candidates (ASCs). We show how to construct such ASCs for numerous settings, including supermodular requirements, Hoffman-Schwartz-type lattice polyhedra, and abstract networks where $π$ fulfils a conservation law. The resulting algorithm can be carried out efficiently when $\mathcal{P}$ and $π$ can be accessed via appropriate oracles. For any system allowing the construction of ASCs, our results imply a simple polyhedral description of the set of marginal vectors for which the decomposition problem is feasible. Finally, we characterize balanced hypergraphs as the systems $(E, \mathcal{P})$ that allow the perfect decomposition of any marginal vector $ρ\in [0,1]^E$, i.e., where we can always find a distribution reaching the highest attainable probability $\operatorname{Pr}[P \cap S \neq \emptyset] = \min \{ \sum_{e \in P} ρ_e, 1\}$ for all $P \in \mathcal{P}$.

cs.DM↗

arXiv2023

Decomposition of Probability Marginals for Security Games in Max-Flow/Min-Cut Systems

Given a set system $(E, \mathcal{P})$ with $ρ\in [0, 1]^E$ and $π\in [0,1]^{ \mathcal{P}}$, our goal is to find a probability distribution for a random set $S \subseteq E$ such that $\operatorname{Pr}[e \in S] = ρ_e$ for all $e \in E$ and $\operatorname{Pr}[P \cap S \neq \emptyset] \geq π_P$ for all $P \in \mathcal{P}$. We extend the results of Dahan, Amin, and Jaillet (MOR 2022) who studied this problem motivated by a security game in a directed acyclic graph (DAG). We focus on the setting where $π$ is of the affine form $π_P = 1 - \sum_{e \in P} μ_e$ for $μ\in [0, 1]^E$. A necessary condition for the existence of the desired distribution is that $\sum_{e \in P} ρ_e \geq π_P$ for all $P \in \mathcal{P}$. We show that this condition is sufficient if and only if $\mathcal{P}$ has the weak max-flow/min-cut property. We further provide an efficient combinatorial algorithm for computing the corresponding distribution in the special case where $(E, \mathcal{P})$ is an abstract network. As a consequence, equilibria for the security game by Dahan et al. can be efficiently computed in a wide variety of settings (including arbitrary digraphs). As a subroutine of our algorithm, we provide a combinatorial algorithm for computing shortest paths in abstract networks, partially answering an open question by McCormick (SODA 1996). We further show that a conservation law proposed by Dahan et al. for the requirement vector $π$ in DAGs can be reduced to the setting of affine requirements described above.

cs.DM↗

arXiv2023

The popular assignment problem: when cardinality is more important than popularity

We consider a matching problem in a bipartite graph $G=(A\cup B,E)$ where nodes in $A$ are agents having preferences in partial order over their neighbors, while nodes in $B$ are objects without preferences. We propose a polynomial-time combinatorial algorithm based on LP duality that finds a maximum matching or assignment in $G$ that is popular among all maximum matchings, if there exists one. Our algorithm can also be used to achieve a trade-off between popularity and cardinality by imposing a penalty on unmatched nodes in $A$. We also provide an $O^*(|E|^k)$ algorithm that finds an assignment whose unpopularity margin is at most $k$; this algorithm is essentially optimal, since the problem is $\mathsf{NP}$-complete and $\mathsf{W}_l[1]$-hard with parameter $k$. We also prove that finding a popular assignment of minimum cost when each edge has an associated binary cost is $\mathsf{NP}$-hard, even if agents have strict preferences. By contrast, we propose a polynomial-time algorithm for the variant of the popular assignment problem with forced/forbidden edges. Finally, we present an application in the context of housing markets.

cs.DS↗

arXiv2022

The TSP with drones: The benefits of retraversing the arcs

In the Traveling Salesman Problem with Drones (TSP-mD), a truck and multiple drones cooperate to serve customers in the minimum amount of time. The drones are launched and retrieved by the truck at customer locations, and each of their flights must not consume more energy than allowed by their batteries. Most problem settings in the literature restrict the feasible truck routes to cycles, i.e., closed paths, which never visit a node more than once. Revisiting a node, however, may lower the time required to serve all the customers. Additionally, we observe that optimal solutions for the TSP-mD may retraverse arcs, i.e., optimal truck routes may contain the same arcs multiple times. We refer to such solutions as arc-retraversing, and include them in our solution space by modeling the truck route as a closed walk. We describe Euclidean instances where all the optimal solutions are arc-retraversing. The necessity of arc retraversals does not seem to have been investigated in previous studies, and those that allow node revisits seem to assume that there always exists an optimal solution without arc retraversals. We prove that under certain conditions, which are commonly met in the literature, this assumption is correct. When these conditions are not met, however, excluding arc-retraversing solutions might result in an increase of the optimal value; we identify cases where a priori and a posteriori upper bounds hold on such increase. Finally, we prove that there is no polynomial-time heuristic that can approximate the metric TSP-mD within a constant factor, unless P=NP. We identify a (non-constant) approximation factor explicitly when the truck can visit all the nodes.

math.OC↗

arXiv2022

Assigning and Scheduling Generalized Malleable Jobs under Subadditive or Submodular Processing Speeds

Malleable scheduling is a model that captures the possibility of parallelization to expedite the completion of time-critical tasks. A malleable job can be allocated and processed simultaneously on multiple machines, occupying the same time interval on all these machines. We study a general version of this setting, in which the functions determining the joint processing speed of machines for a given job follow different discrete concavity assumptions (subadditivity, fractional subadditivity, submodularity, and matroid ranks). We show that under these assumptions the problem of scheduling malleable jobs at minimum makespan can be approximated by a considerably simpler assignment problem. Moreover, we provide efficient approximation algorithms for both the scheduling and the assignment problem, with increasingly stronger guarantees for increasingly stronger concavity assumptions, including a logarithmic approximation factor for the case of submodular processing speeds and a constant approximation factor when processing speeds are determined by matroid rank functions. Computational experiments indicate that our algorithms outperform the theoretical worst-case guarantees.

cs.DM↗

arXiv2021

Multi-Leader Congestion Games with an Adversary

We study a multi-leader single-follower congestion game where multiple users (leaders) choose one resource out of a set of resources and, after observing the realized loads, an adversary (single-follower) attacks the resources with maximum loads, causing additional costs for the leaders. For the resulting strategic game among the leaders, we show that pure Nash equilibria may fail to exist and therefore, we consider approximate equilibria instead. As our first main result, we show that the existence of a $K$-approximate equilibrium can always be guaranteed, where $K \approx 1.1974$ is the unique solution of a cubic polynomial equation. To this end, we give a polynomial time combinatorial algorithm which computes a $K$-approximate equilibrium. The factor $K$ is tight, meaning that there is an instance that does not admit an $α$-approximate equilibrium for any $α<K$. Thus $α=K$ is the smallest possible value of $α$ such that the existence of an $α$-approximate equilibrium can be guaranteed for any instance of the considered game. Secondly, we focus on approximate equilibria of a given fixed instance. We show how to compute efficiently a best approximate equilibrium, that is, with smallest possible $α$ among all $α$-approximate equilibria of the given instance.

cs.GT↗

arXiv2021

A Constant-Factor Approximation for Generalized Malleable Scheduling under $M^\natural$-Concave Processing Speeds

In generalized malleable scheduling, jobs can be allocated and processed simultaneously on multiple machines so as to reduce the overall makespan of the schedule. The required processing time for each job is determined by the joint processing speed of the allocated machines. We study the case that processing speeds are job-dependent $M^\natural$-concave functions and provide a constant-factor approximation for this setting, significantly expanding the realm of functions for which such an approximation is possible. Further, we explore the connection between malleable scheduling and the problem of fairly allocating items to a set of agents with distinct utility functions, devising a black-box reduction that allows to obtain resource-augmented approximation algorithms for the latter.

cs.DM↗

arXiv2021

Bifactor Approximation for Location Routing with Vehicle and Facility Capacities

Location Routing is a fundamental planning problem in logistics, in which strategic location decisions on the placement of facilities (depots, distribution centers, warehouses etc.) are taken based on accurate estimates of operational routing costs. We present an approximation algorithm, i.e., an algorithm with proven worst-case guarantees both in terms of running time and solution quality, for the general capacitated version of this problem, in which both vehicles and facilities are capacitated. Before, such algorithms were only known for the special case where facilities are uncapacitated or where their capacities can be extended arbitrarily at linear cost. Previously established lower bounds that are known to approximate the optimal solution value well in the uncapacitated case can be off by an arbitrary factor in the general case. We show that this issue can be overcome by a bifactor approximation algorithm that may slightly exceed facility capacities by an adjustable, arbitrarily small margin while approximating the optimal cost by a constant factor. In addition to these proven worst-case guarantees, we also assess the practical performance of our algorithm in a comprehensive computational study, showing that the approach allows efficient computation of near-optimal solutions for instance sizes beyond the reach of current state-of-the-art heuristics.

cs.DM↗

arXiv2020

Single-machine scheduling with an external resource

This paper studies the complexity of single-machine scheduling with an external resource, which is rented for a non-interrupted period. Jobs that need this external resource are executed only when the external resource is available. There is a cost associated with the scheduling of jobs and a cost associated with the duration of the renting period of the external resource. We look at four classes of problems with an external resource: a class of problems where the renting period is budgeted and the scheduling cost needs to be minimized, a class of problems where the scheduling cost is budgeted and the renting period needs to be minimized, a class of two-objective problems where both, the renting period and the scheduling cost, are to be minimized, and a class of problems where a linear combination of the scheduling cost and the renting period is minimized. We provide a thorough complexity analysis (NP-hardness proofs and (pseudo-)polynomial algorithms) for different members of these four classes.

cs.DS↗

arXiv2020

Pure Nash Equilibria in Resource Graph Games

This paper studies the existence of pure Nash equilibria in resource graph games, which are a general class of strategic games used to succinctly represent the players' private costs. There is a finite set of resources and the strategy set of each player corresponds to a set of subsets of resources. The cost of a resource is an arbitrary function that depends on the load vector of the resources in a specified neighborhood. As our main result, we give complete characterizations of the cost functions guaranteeing the existence of pure Nash equilibria for weighted and unweighted players, respectively. 1. For unweighted players, pure Nash equilibria are guaranteed to exist for any choice of the players' strategy space if and only if the cost of each resource is an arbitrary function of the load of the resource itself and linear in the load of all other resources where the linear coefficients of mutual influence of different resources are symmetric. 2. For games with weighted players, pure Nash equilibria are guaranteed to exist for any choice of the players' strategy space if and only if the cost of a resource is linear in all resource loads, and the linear factors of mutual influence are symmetric, or there is no interaction among resources and the cost is an exponential function of the local resource load. 3. For the special case that the players' strategy sets are matroids, we show that pure Nash equilibria exist under a local monotonicity property, even when cost functions are player-specific. We point out an application of this result to bilevel load balancing games, which are motivated by the study of network infrastructures that are resilient against external attackers and internal congestion effects. 4. Finally, we derive hardness results for deciding whether a given strategy is a pure Nash equilibrium for network routing games and matroid games, respectively.

cs.GT↗