CyberSec Research

Any Littlestone class, or stable graph, has finite sets which function as ``virtual elements'': these can be seen from the learning side as representing hypotheses which are expressible as weighted majority opinions of hypotheses in the class, and from the model-theoretic side as an approximate finitary version of realizing types. We introduce and study the epsilon-saturation of a Littlestone class, or stable graph, which is essentially the closure of the class under inductively adding all such virtual elements. We characterize this closure and prove that under reasonable choices of parameters, it remains Littlestone (or stable), though not always of the same Littlestone dimension. This highlights some surprising phenomena having to do with regimes of epsilon and the relation between Littlestone/stability and VC dimension.

It is known that any two trees on the same $n$ leaves can be displayed by a network with $n-2$ reticulations, and there are two trees that cannot be displayed by a network with fewer reticulations. But how many reticulations are needed to display multiple trees? For any set of $t$ trees on $n$ leaves, there is a trivial network with $(t - 1)n$ reticulations that displays them. To do better, we have to exploit common structure of the trees to embed non-trivial subtrees of different trees into the same part of the network. In this paper, we show that for $t \in o(\sqrt{\lg n})$, there is a set of $t$ trees with virtually no common structure that could be exploited. More precisely, we show for any $t\in o(\sqrt{\lg n})$, there are $t$ trees such that any network displaying them has $(t-1)n - o(n)$ reticulations. For $t \in o(\lg n)$, we obtain a slightly weaker bound. We also prove that already for $t = c\lg n$, for any constant $c > 0$, there is a set of $t$ trees that cannot be displayed by a network with $o(n \lg n)$ reticulations, matching up to constant factors the known upper bound of $O(n \lg n)$ reticulations sufficient to display \emph{all} trees with $n$ leaves. These results are based on simple counting arguments and extend to unrooted networks and trees.

The \emph{signed series} problem in the $\ell_2$ norm asks, given set of vectors $v_1,\ldots,v_n\in \mathbf{R}^d$ having at most unit $\ell_2$ norm, does there always exist a series $(\varepsilon_i)_{i\in [n]}$ of $\pm 1$ signs such that for all $i\in [n]$, $\max_{i\in [n]} \|\sum_{j=1}^i \varepsilon_i v_i\|_2 = O(\sqrt{d})$. A result of Banaszczyk [2012, \emph{Rand. Struct. Alg.}] states that there exist signs $\varepsilon_i\in \{-1,1\},\; i\in [n]$ such that $\max_{i\in [n]} \|\sum_{j=1}^i \varepsilon_i v_i\|_2 = O(\sqrt{d+\log n})$. The best constructive bound known so far is of $O(\sqrt{d\log n})$, by Bansal and Garg [2017, \emph{STOC.}, 2019, \emph{SIAM J. Comput.}]. We give a polynomial-time randomized algorithm to find signs $x(i) \in \{-1,1\},\; i\in [n]$ such that \[ \max_{i\in [n]} \|\sum_{j=1}^i x(i)v_i\|_2 = O(\sqrt{d + \log^2 n}) = O(\sqrt{d}+\log n).\] By the constructive reduction of Harvey and Samadi [\emph{COLT}, 2014], this also yields a constructive bound of $O(\sqrt{d}+\log n)$ for the Steinitz problem in the $\ell_2$-norm. Thus, our result settles both conjectures when $d \geq \log^2n$. Our algorithm is based on the framework on Bansal and Garg, together with a new analysis involving $(i)$ additional linear and spectral orthogonality constraints during the construction of the covariance matrix of the random walk steps, which allow us to control the quadratic variation in the linear as well as the quadratic components of the discrepancy increment vector, alongwith $(ii)$ a ``Freedman-like" version of the Hanson-Wright concentration inequality, for filtration-dependent sums of subgaussian chaoses.

We prove that the signed counting (with respect to the parity of the ``$\operatorname{inv}$'' statistic) of partition matrices equals the cardinality of a subclass of inversion sequences. In the course of establishing this result, we introduce an interesting class of partition matrices called improper partition matrices. We further show that a subset of improper partition matrices is equinumerous with the set of Motzkin paths. Such an equidistribution is established both analytically and bijectively.

Ransomware impact hinges on how easily an intruder can move laterally and spread to the maximum number of assets. We present a graph-theoretic method to measure lateral-movement susceptibility and estimate blast radius. We build a directed multigraph where vertices represent assets and edges represent reachable services (e.g., RDP/SSH) between them. We model lateral movement as a probabilistic process using a pivot potential factor $\pi(s)$ for each service. This allows us to iteratively compute a $K$-hop compromise probability matrix that captures how compromise propagates through the network. Metrics derived from this model include: (1) Lateral-Movement Susceptibility (LMS$_K$): the average probability of a successful lateral movement between any two assets (0-1 scale); and (2) Blast-Radius Estimate (BRE$_K$): the expected percentage of assets compromised in an average attack scenario. Interactive control (SSH 22, RDP 3389) gets higher $\pi(s)$ than app-only ports (MySQL 3306, MSSQL 1433), which seldom enable pivoting without an RCE. Across anonymized enterprise snapshots, pruning high-$\pi(s)$ edges yields the largest LMS$_K$/BRE$_K$ drop, aligning with CISA guidance, MITRE ATT\&CK (TA0008: Lateral Movement), and NIST SP~800-207. The framework evaluates (micro)segmentation and helps prioritize controls that reduce lateral movement susceptibility and shrink blast radius.

Workforce scheduling in the healthcare sector is a significant operational challenge, characterized by fluctuating patient loads, diverse clinical skills, and the critical need to control labor costs while upholding high standards of patient care. This problem is inherently multi-objective, demanding a delicate balance between competing goals: minimizing payroll, ensuring adequate staffing for patient needs, and accommodating staff preferences to mitigate burnout. We propose a Multi-objective Genetic Algorithm (MOO-GA) that models the hospital unit workforce scheduling problem as a multi-objective optimization task. Our model incorporates real-world complexities, including hourly appointment-driven demand and the use of modular shifts for a multi-skilled workforce. By defining objective functions for cost, patient care coverage, and staff satisfaction, the GA navigates the vast search space to identify a set of high-quality, non-dominated solutions. Demonstrated on datasets representing a typical hospital unit, the results show that our MOO-GA generates robust and balanced schedules. On average, the schedules produced by our algorithm showed a 66\% performance improvement over a baseline that simulates a conventional, manual scheduling process. This approach effectively manages trade-offs between critical operational and staff-centric objectives, providing a practical decision support tool for nurse managers and hospital administrators.

Let $\mathcal{F}$ be a family of graphs. A graph is said to be $\mathcal{F}$-free if it contains no member of $\mathcal{F}$. The generalized Tur\'{a}n number $ex(n,H,\mathcal{F})$ denotes the maximum number of copies of a graph $H$ in an $n$-vertex $\mathcal{F}$-free graph, while the generalized edge Tur\'{a}n number $mex(m,H,\mathcal{F})$ denotes the maximum number of copies of $H$ in an $m$-edge $\mathcal{F}$-free graph. It is well known that if a graph has maximum degree $d$, then it is $K_{1,d+1}$-free. Wood \cite{wood} proved that $ex(n,K_t,K_{1,d+1}) \leq \frac{n}{d+1}\binom{d+1}{t}$. More recently, Chakraborty and Chen \cite{CHAKRABORTI2024103955} established analogous bounds for graphs with bounded maximum path length: $mex(m,K_t,P_{r+1}) \leq \frac{m}{\binom{r}{2}}\binom{r}{t}$. In this paper, we improve these bounds using the localization technique, based on suitably defined local parameters. Furthermore, we characterize the extremal graphs attaining these improved bounds.

Let $\mathcal{F}$ be a family of graphs. A graph is called $\mathcal{F}$-free if it does not contain any member of $\mathcal{F}$. Generalized Tur\'{a}n problems aim to maximize the number of copies of a graph $H$ in an $n$-vertex $\mathcal{F}$-free graph. This maximum is denoted by $ex(n, H, \mathcal{F})$. When $H \cong K_2$, it is simply denoted by $ex(n,F)$. Erd\H{o}s and Gallai established the bounds $ex(n, P_{k+1}) \leq \frac{n(k-1)}{2}$ and $ex(n, C_{\geq k+1}) \leq \frac{k(n-1)}{2}$. This was later extended by Luo \cite{luo2018maximum}, who showed that $ex(n, K_s, P_{k+1}) \leq \frac{n}{k} \binom{k}{s}$ and $ex(n, K_s, C_{\geq k+1}) \leq \frac{n-1}{k-1} \binom{k}{s}$. Let $N(G,K_s)$ denote the number of copies of $K_s$ in $G$. In this paper, we use the vertex-based localization framework, introduced in \cite{adak2025vertex}, to generalize Luo's bounds. In a graph $G$, for each $v \in V(G)$, define $p(v)$ to be the length of the longest path that contains $v$. We show that \[N(G,K_s) \leq \sum_{v \in V(G)} \frac{1}{p(v)+1}{p(v)+1\choose s} = \frac{1}{s}\sum_{v \in V(G)}{p(v) \choose s-1}\] We strengthen the cycle bound from \cite{luo2018maximum} as follows: In graph $G$, for each $v \in V(G)$, let $c(v)$ be the length of the longest cycle that contains $v$, or $2$ if $v$ is not part of any cycle. We prove that \[N(G,K_s) \leq \left(\sum_{v\in V(G)}\frac{1}{c(v)-1}{c(v) \choose s}\right) - \frac{1}{c(u)-1}{c(u) \choose s}\] where $c(u)$ denotes the circumference of $G$. We provide full proofs for the cases $s = 1$ and $s \geq 3$, while the case $s = 2$ follows from the result in \cite{adak2025vertex}. \newline Furthermore, we characterize the class of extremal graphs that attain equality for these bounds.

For $0\leq \rho\leq 1$, a $\rho$-happy vertex $v$ in a coloured graph shares colour with at least $\rho\mathrm{deg}(v)$ of its neighbours. Soft happy colouring of a graph $G$ with $k$ colours extends a partial $k$-colouring to a complete vertex $k$-colouring such that the number of $\rho$-happy vertices is maximum among all such colouring extensions. The problem is known to be NP-hard, and an optimal solution has a direct relation with the community structure of the graph. In addition, some heuristics and local search algorithms, such as {\sf Local Maximal Colouring} ({\sf LMC}) and {\sf Local Search} ({\sf LS}), have already been introduced in the literature. In this paper, we design Genetic and Memetic Algorithms for soft happy colouring and test them for a large set of randomly generated partially coloured graphs. Memetic Algorithms yield a higher number of $\rho$-happy vertices, but Genetic Algorithms can perform well only when their initial populations are locally improved by {\sf LMC} or {\sf LS}. Statistically significant results indicate that both Genetic and Memetic Algorithms achieve high average accuracy in community detection when their initial populations are enhanced using {\sf LMC}. Moreover, among the competing methods, the evolutionary algorithms identified the greatest number of complete solutions.

We prove that for every integer $n > 0$ and for every alphabet $\Sigma_k$ of size $k \geq 3$, there exists a necklace of length $n$ whose Burrows-Wheeler Transform (BWT) is completely unclustered, i.e., it consists of exactly $n$ runs with no two consecutive equal symbols. These words represent the worst-case behavior of the BWT for clustering, since the number of BWT runs is maximized. We also establish a lower bound on their number. This contrasts with the binary case, where the existence of infinitely many completely unclustered BWTs is still an open problem, related to Artin's conjecture on primitive roots.

Computers and algorithms play an ever-increasing role in obtaining new results in graph theory. In this survey, we present a broad range of techniques used in computer-assisted graph theory, including the exhaustive generation of all pairwise non-isomorphic graphs within a given class, the use of searchable databases containing graphs and invariants as well as other established and emerging algorithmic paradigms. We cover approaches based on mixed integer linear programming, semidefinite programming, dynamic programming, SAT solving, metaheuristics and machine learning. The techniques are illustrated with numerous detailed results covering several important subareas of graph theory such as extremal graph theory, graph coloring, structural graph theory, spectral graph theory, regular graphs, topological graph theory, special sets in graphs, algebraic graph theory and chemical graph theory. We also present some smaller new results that demonstrate how readily a computer-assisted graph theory approach can be applied once the appropriate tools have been developed.

We study the algorithmic problem of finding large $\gamma$-balanced independent sets in dense random bipartite graphs; an independent set is $\gamma$-balanced if a $\gamma$ proportion of its vertices lie on one side of the bipartition. In the sparse regime, Perkins and Wang established tight bounds within the low-degree polynomial (LDP) framework, showing a factor-$1/(1-\gamma)$ statistical-computational gap via the Overlap Gap Property (OGP) framework tailored for stable algorithms. However, these techniques do not appear to extend to the dense setting. For the related large independent set problem in dense random graph, the best known algorithm is an online greedy procedure that is inherently unstable, and LDP algorithms are conjectured to fail even in the "easy" regime where greedy succeeds. We show that the largest $\gamma$-balanced independent set in dense random bipartite graphs has size $\alpha:=\frac{\log_b n}{\gamma(1-\gamma)}$ whp, where $n$ is the size of each bipartition, $p$ is the edge probability, and $b=1/(1-p)$. We design an online algorithm that achieves $(1-\epsilon)(1-\gamma)\alpha$ whp for any $\epsilon>0$. We complement this with a sharp lower bound, showing that no online algorithm can achieve $(1+\epsilon)(1-\gamma)\alpha$ with nonnegligible probability. Our results suggest that the same factor-$1/(1-\gamma)$ gap is also present in the dense setting, supporting its conjectured universality. While the classical greedy procedure on $G(n,p)$ is straightforward, our algorithm is more intricate: it proceeds in two stages, incorporating a stopping time and suitable truncation to ensure that $\gamma$-balancedness-a global constraint-is met despite operating with limited information. Our lower bound utilizes the OGP framework; we build on a recent refinement of this framework for online models and extend it to the bipartite setting.

Generally, networks are classified into two sides of inequality and equality with respect to the number of links at nodes by the types of degree distributions. One side includes many social, technological, and biological networks which consist of a few nodes with many links, and many nodes with a few links, whereas the other side consists of all nodes with an equal number of links. In comprehensive investigations between them, we have found that, as a more equal network, the tolerance of whole connectivity is stronger without fragmentation against the malfunction of nodes in a wide class of randomized networks. However, we newly find that all networks which include typical well-known network structures between them become extremely vulnerable, if a strong modular (or community) structure is added with commonalities of areas, interests, religions, purpose, and so on. These results will encourage avoiding too dense unions by connecting nodes and taking into account the balanced resource allocation between intra- and inter-links of weak communities. We must reconsider not only efficiency but also tolerance against attacks or disasters, unless no community that is really impossible.

The growth form (or corona limit) of a tiling is the limit form of its coordination shells, i.e. its set of tiles located at a fixed distance from some tile. We give an overview of current results, conjectures and open questions about growth forms, including periodic, multigrid, substitution, and hat tilings.

A well-known result by Kant [Algorithmica, 1996] implies that n-vertex outerplane graphs admit embedding-preserving planar straight-line grid drawings where the internal faces are convex polygons in $O(n^2)$ area. In this paper, we present an algorithm to compute such drawings in $O(n^{1.5})$ area. We also consider outerplanar drawings in which the internal faces are required to be strictly-convex polygons. In this setting, we consider outerplanar graphs whose weak dual is a path and give a drawing algorithm that achieves $\Theta(nk^2)$ area, where $k$ is the maximum size of an internal facial cycle.

"Metaphorical maps" or "contact representations" are visual representations of vertex-weighted graphs that rely on the geographic map metaphor. The vertices are represented by countries, the weights by the areas of the countries, and the edges by contacts/ boundaries among them. The accuracy with which the weights are mapped to areas and the simplicity of the polygons representing the countries are the two classical optimization goals for metaphorical maps. Mchedlidze and Schnorr [Metaphoric Maps for Dynamic Vertex-weighted Graphs, EuroVis 2022] presented a force-based algorithm that creates metaphorical maps that balance between these two optimization goals. Their maps look visually simple, but the accuracy of the maps is far from optimal - the countries' areas can vary up to 30% compared to required. In this paper, we provide a multi-fold extension of the algorithm in [Metaphoric Maps for Dynamic Vertex-weighted Graphs, EuroVis 2022]. More specifically: 1. Towards improving accuracy: We introduce the notion of region stiffness and suggest a technique for varying the stiffness based on the current pressure of map regions. 2. Towards maintaining simplicity: We introduce a weight coefficient to the pressure force exerted on each polygon point based on whether the corresponding point appears along a narrow passage. 3. Towards generality: We cover, in contrast to [Metaphoric Maps for Dynamic Vertex-weighted Graphs, EuroVis 2022], non-triangulated graphs. This is done by either generating points where more than three regions meet or by introducing holes in the metaphorical map. We perform an extended experimental evaluation that, among other results, reveals that our algorithm is able to construct metaphorical maps with nearly perfect area accuracy with a little sacrifice in their simplicity.

The domatic number of a graph $G$ is the maximum number of pairwise disjoint dominating sets of $G$. We are interested in the LP-relaxation of this parameter, which is called the fractional domatic number of $G$. We study its extremal value in the class of graphs of minimum degree $d$. The fractional domatic number of a graph of minimum degree $d$ is always at most $d+1$, and at least $(1-o(1))\, d/\ln d$ as $d\to \infty$. This is asymptotically tight even within the class of split graphs. Our main result concerns the case $d=2$; we show that, excluding $8$ exceptional graphs, the fractional domatic number of every connected graph of minimum degree (at least) $2$ is at least $5/2$. We also show that this bound cannot be improved if only finitely many graphs are excluded, even when restricting to bipartite graphs of girth at least $6$. This proves in a stronger sense a conjecture by Gadouleau, Harms, Mertzios, and Zamaraev (2024). This also extends and generalises results from McCuaig and Shepherd (1989), from Fujita, Kameda, and Yamashita (2000), and from Abbas, Egerstedt, Liu, Thomas, and Whalen (2016). Finally, we show that planar graphs of minimum degree at least $2$ and girth at least $g$ have fractional domatic number at least $3 - O(1/g)$ as $g\to\infty$.

We study the asymptotic size of the Karp-Sipser core in the configuration model with arbitrary degree distributions. The Karp-Sipser core is the induced subgraph obtained by iteratively removing all leaves and their neighbors through the leaf-removal process, and finally discarding any isolated vertices \cite{BCC}. Our main result establishes the convergence of the Karp-Sipser core size to an explicit fixed-point equation under general degree assumptions.The approach is based on analyzing the corresponding local weak limit of the configuration model - a unimodular Galton-Watson tree and tracing the evolution process of all vertex states under leaf-removal dynamics by use of the working mechanism of an enhanced version of Warning Propagation along with Node Labeling Propagation.