Searcharxiv⌕ Search

arXiv subjects

Hyunwoo Lee

Publications and source records attributed to Hyunwoo Lee.

At least 37 records · Page 2Linked to original sources

Random Access for LEO Satellite Communication Systems via Deep Learning

Integrating contention-based random access procedures into low Earth orbit (LEO) satellite communication (SatCom) systems poses new challenges, including long propagation delays, large Doppler shifts, and a large number of simultaneous access attempts. These factors degrade the efficiency and responsiveness of conventional random access schemes, particularly in scenarios such as satellite-based internet of things and direct-to-device services. In this paper, we propose a deep learning-based random access framework designed for LEO SatCom systems. The framework incorporates an early preamble collision classifier that uses multi-antenna correlation features and a lightweight 1D convolutional neural network to estimate the number of collided users at the earliest stage. Based on this estimate, we introduce an opportunistic transmission scheme that balances access probability and resource efficiency to improve success rates and reduce delay. Simulation results under 3GPP-compliant LEO settings confirm that the proposed framework achieves higher access success probability, lower delay, better physical uplink shared channel utilization, and reduced computational complexity compared to existing schemes.

eess.SP↗

Continuous-Time Analysis for Minimax and Bilevel Problems

We study single-loop gradient-flow dynamics for nested optimization, where the outer variable evolves while auxiliary variables track the inner solution map. While existing analyses typically rely on problem- and condition-specific Lyapunov constructions, we propose, to our knowledge, the first unified Lyapunov template for continuous-time analysis that covers minimax, bilevel via a lifted penalty formulation, and min--min--max. Our proof is modular, built from reusable lemmas that yield a unified characterization of time-scale separation. This characterization bridges regimes from strong convexity/concavity to mere convexity through an error-bound condition, and produces explicit closed-form thresholds that avoid the coupled ratio conditions common in discrete-time analyses. We further compare the penalty dynamics with the ideal hyper-gradient flow, derive a finite-time tracking bound, and discuss an Euler one-step analogue; hypercleaning diagnostics show that the predicted relative time-scale regions remain visible under stable forward-Euler discretization.

math.OC↗

Random-Restart Best-Response Dynamics for Large-Scale Integer Programming Games and Their Applications

This paper presents scalable algorithms for computing pure Nash equilibria (PNEs) in large-scale integer programming games (IPGs), where existing exact methods typically handle only small numbers of players. Motivated by a county-level aquatic invasive species (AIS) prevention problem with 84 decision makers, we develop and analyze random-restart best-response dynamics (RR-BRD), a randomized search framework for PNEs. For IPGs with finite action sets, we model RR-BRD as a Markov chain on the best-response state graph and show that, whenever a PNE exists and the restart law has positive probability of reaching a PNE within the round cap, RR-BRD finds a PNE almost surely. We also propose a Monte Carlo sampling-and-simulation procedure to estimate success behavior under a fixed round cap, which informs our instance-dependent performance characterization. We then embed RR-BRD as a randomized local-search subroutine within the zero-regret (ZR) framework, yielding BRD-incorporated zero-regret (BZR). Using solver callbacks, RR-BRD searches for and supplies PNEs, while ZR separates and adds equilibrium inequalities to tighten the formulation. We introduce edge-weighted budgeted maximum coverage (EBMC) games to model AIS prevention and establish PNE existence results for both selfish and locally altruistic utilities. Computational experiments on synthetic EBMC and knapsack problem game instances show that RR-BRD and BZR scale equilibrium computation up to $n \le 30$ players. We further solve a real-world EBMC game derived from the Minnesota AIS dataset with $n = 84$ county players.

cs.GT↗

On a weaker notion of cross $t$-intersecting families

We prove that if two families $\mathcal{F} \subseteq \binom{[n]}{k}$ and $\mathcal{F}' \subseteq \binom{[n]}{k'}$ satisfy $\sum_{1 \leq i, j \leq \ell} \lvert F_i \cap F_j' \rvert \geq \ell^2t - \ell +1$ for every choice of distinct $F_1, \ldots, F_\ell \in \mathcal{F}$ and $F_1', \ldots, F_\ell' \in \mathcal{F}'$, then $\lvert \mathcal{F} \rvert \cdot \lvert \mathcal{F}' \rvert \leq \binom{n-t}{k-t} \binom{n-t}{k'-t}$, provided that $n$ is sufficiently large. This extends a celebrated theorem of Pyber for large $n$, which determines the tight upper bound for the product of the sizes of cross $1$-intersecting families.

math.CO↗

Anticoncentration of random spanning trees in almost regular graphs

The celebrated formula of Otter \emph{[Ann. of Math. (2) 49 (1948), 583--599]} asserts that the complete graph contains exponentially many non-isomorphic spanning trees. In this paper, we show that every connected almost regular graph with sufficiently large degree already contains exponentially many non-isomorphic spanning trees. Indeed, we prove a stronger statement: for every fixed $n$-vertex tree $T$, $$ \Pr\bigl[\mathcal{T} \simeq_{\mathrm{iso}} T\bigr] = e^{-Ω(n)}, $$ where $\mathcal{T}$ is a uniformly random spanning tree of a connected $n$-vertex almost regular graph with sufficiently large degree. To prove this, we introduce a graph-theoretic variant of the classical balls--into--bins model, which may be of independent interest.

math.CO↗

Beyond Gaussian Initializations: Signal Preserving Weight Initialization for Odd-Sigmoid Activations

Activation functions critically influence trainability and expressivity, and recent work has therefore explored a broad range of nonlinearities. However, widely used Gaussian i.i.d. initializations are designed to preserve activation variance under wide or infinite width assumptions. In deep and relatively narrow networks with sigmoidal nonlinearities, these schemes often drive preactivations into saturation, and collapse gradients. To address this, we introduce an odd-sigmoid activations and propose an activation aware initialization tailored to any function in this class. Our method remains robust over a wide band of variance scales, preserving both forward signal variance and backpropagated gradient norms even in very deep and narrow networks. Empirically, across standard image benchmarks we find that the proposed initialization is substantially less sensitive to depth, width, and activation scale than Gaussian initializations. In physics informed neural networks (PINNs), scaled odd-sigmoid activations combined with our initialization achieve lower losses than Gaussian based setups, suggesting that diagonal-plus-noise weights provide a practical alternative when Gaussian initialization breaks down.

cs.LG↗

On Sidorenko exponents of hypergraphs

For an $r$-graph $F$, define Sidorenko exponent $s(F)$ as $$s(F):= \sup \{s \geq 0: \exists \text{$r$-graph $H$ s.t. } t_F(H) = t_{K^{(r)}_r} (H)^s > 0\},$$ where $t_{H_1}(H_2)$ denotes the homomorphism density of $H_1$ in $H_2$. The celebrated Sidorenko's conjecture states that $s(F) = e(F)$ holds for every bipartite graph $F$. It is known that for all $r \geq 3$, the $r$-uniform version of Sidorenko's conjecture is false, and only a few hypergraphs are known to be Sidorenko. In this paper, we discover a new broad class of Sidorenko hypergraphs and obtain general upper bounds on $s(F)$ for certain hypergraphs related to dominating hypergraphs. This makes progress toward a problem raised by Nie and Spiro. We also discover a new connection between Sidorenko exponents and upper bounds on the extremal numbers of a large class of hypergraphs, which generalizes the hypergraph analogue of Kővári--Sós--Turán theorem proved by Erdős.

math.CO↗

Mitigating Distribution Shift in Stock Price Data via Return-Volatility Normalization for Accurate Prediction

How can we address distribution shifts in stock price data to improve stock price prediction accuracy? Stock price prediction has attracted attention from both academia and industry, driven by its potential to uncover complex market patterns and enhance decisionmaking. However, existing methods often fail to handle distribution shifts effectively, focusing on scaling or representation adaptation without fully addressing distributional discrepancies and shape misalignments between training and test data. We propose ReVol (Return-Volatility Normalization for Mitigating Distribution Shift in Stock Price Data), a robust method for stock price prediction that explicitly addresses the distribution shift problem. ReVol leverages three key strategies to mitigate these shifts: (1) normalizing price features to remove sample-specific characteristics, including return, volatility, and price scale, (2) employing an attention-based module to estimate these characteristics accurately, thereby reducing the influence of market anomalies, and (3) reintegrating the sample characteristics into the predictive process, restoring the traits lost during normalization. Additionally, ReVol combines geometric Brownian motion for long-term trend modeling with neural networks for short-term pattern recognition, unifying their complementary strengths. Extensive experiments on real-world datasets demonstrate that ReVol enhances the performance of the state-of-the-art backbone models in most cases, achieving an average improvement of more than 0.03 in IC and over 0.7 in SR across various settings.

q-fin.ST↗

VWAttacker: A Systematic Security Testing Framework for Voice over WiFi User Equipments

We present VWAttacker, the first systematic testing framework for analyzing the security of Voice over WiFi (VoWiFi) User Equipment (UE) implementations. VWAttacker includes a complete VoWiFi network testbed that communicates with Commercial-Off-The-Shelf (COTS) UEs based on a simple interface to test the behavior of diverse VoWiFi UE implementations; uses property-guided adversarial testing to uncover security issues in different UEs systematically. To reduce manual effort in extracting and testing properties, we introduce an LLM-based, semi-automatic, and scalable approach for property extraction and testcase (TC) generation. These TCs are systematically mutated by two domain-specific transformations. Furthermore, we introduce two deterministic oracles to detect property violations automatically. Coupled with these techniques, VWAttacker extracts 63 properties from 11 specifications, evaluates 1,116 testcases, and detects 13 issues in 21 UEs. The issues range from enforcing a DH shared secret to 0 to supporting weak algorithms. These issues result in attacks that expose the victim UE's identity or establish weak channels, thus severely hampering the security of cellular networks. We responsibly disclose the findings to all the related vendors. At the time of writing, one of the vulnerabilities has been acknowledged by MediaTek with high severity.

cs.CR↗

On a Ramsey--Turán variant of Roth's theorem

A classical theorem of Roth states that the maximum size of a solution-free set of a homogeneous linear equation $\mathcal{L}$ in $\mathbb{F}_p$ is $o(p)$ if and only if the sum of the coefficients of $\mathcal{L}$ is $0$. In this paper, we prove a Ramsey--Turán variant of Roth's theorem, with respect to a natural notion of ``structured'' sets introduced by Erdős and Sárközy in the 1970's. Namely, we show that the following statements are equivalent: $(a)$ Every solution-free set $A$ of $\mathcal{L}$ in $\mathbb{F}_p$ with $α(\mathrm{Cay}_{\mathbb{F}_p}(A)) = o(p)$ has size $o(p)$. $(b)$ There exists a non-empty \emph{subset} of coefficients of $\mathcal{L}$ with zero sum.

math.CO↗

ZeroDL: Zero-shot Distribution Learning for Text Clustering via Large Language Models

The advancements in large language models (LLMs) have brought significant progress in NLP tasks. However, if a task cannot be fully described in prompts, the models could fail to carry out the task. In this paper, we propose a simple yet effective method to contextualize a task toward a LLM. The method utilizes (1) open-ended zero-shot inference from the entire dataset, (2) aggregate the inference results, and (3) finally incorporate the aggregated meta-information for the actual task. We show the effectiveness in text clustering tasks, empowering LLMs to perform text-to-text-based clustering and leading to improvements on several datasets. Furthermore, we explore the generated class labels for clustering, showing how the LLM understands the task through data.

cs.CL↗

Taxonomy and Analysis of Sensitive User Queries in Generative AI Search

Although there has been a growing interest among industries in integrating generative LLMs into their services, limited experience and scarcity of resources act as a barrier in launching and servicing large-scale LLM-based services. In this paper, we share our experiences in developing and operating generative AI models within a national-scale search engine, with a specific focus on the sensitiveness of user queries. We propose a taxonomy for sensitive search queries, outline our approaches, and present a comprehensive analysis report on sensitive queries from actual users. We believe that our experiences in launching generative AI search systems can contribute to reducing the barrier in building generative LLM-based services.

cs.IR↗

On the order of intersecting hypergraphs

Determining the maximum number of edges in an intersecting hypergraph on a fixed ground set under additional constraints is one of the central topics in extremal combinatorics. In contrast, there are few results on analogous problems concerning the maximum order of such hypergraphs. In this paper, we systematically study these vertex analogues.

math.CO↗

Spanning subdivisions in dense digraphs

We prove that an $n$-vertex digraph $D$ with minimum semi-degree at least $\left(\frac{1}{2} + \varepsilon \right)n$ and $n \geq C m$ contains a subdivision of all $m$-arc digraphs without isolated vertices. Here, $C$ is a constant only depending on $\varepsilon.$ This is the best possible and settles a conjecture raised by Pavez-Signé in a stronger form.

math.CO↗

Towards a high-dimensional Dirac's theorem

Dirac's theorem determines the sharp minimum degree threshold for graphs to contain perfect matchings and Hamiltonian cycles. There have been various attempts to generalize this theorem to hypergraphs with larger uniformity by considering hypergraph matchings and Hamiltonian cycles. In this paper, we consider another natural generalization of perfect matchings, Steiner triple systems. As a Steiner triple system can be viewed as a partition of pairs of vertices, it is a natural high-dimensional analogue of a perfect matching in graphs. We prove that for sufficiently large integer $n$ with $n \equiv 1 \text{ or } 3 \pmod{6}$, any $n$-vertex $3$-uniform hypergraph $H$ with minimum codegree at least $\left(\frac{3 + \sqrt{57}}{12} + o(1) \right)n = (0.879... + o(1))n$ contains a Steiner triple system. In fact, we prove a stronger statement by considering transversal Steiner triple systems in a collection of hypergraphs. We conjecture that the number $\frac{3 + \sqrt{57}}{12}$ can be replaced with $\frac{3}{4}$ which would provide an asymptotically tight high-dimensional generalization of Dirac's theorem.

math.CO↗

Dirac's theorem for linear hypergraphs

Dirac's theorem states that any $n$-vertex graph $G$ with even integer $n$ satisfying $δ(G) \geq n/2$ contains a perfect matching. We generalize this to $k$-uniform linear hypergraphs by proving the following. Any $n$-vertex $k$-uniform linear hypergraph $H$ with minimum degree at least $\frac{n}{k} + Ω(1)$ contains a matching that covers at least $(1-o(1))n$ vertices. This minimum degree condition is asymptotically tight and obtaining a perfect matching is impossible with any degree condition. Furthermore, we show that if $δ(H) \geq (\frac{1}{k}+o(1))n$, then $H$ contains almost spanning linear cycles, almost spanning hypertrees with $o(n)$ leaves, and ``long subdivisions'' of any $o(\sqrt{n})$-vertex graphs.

math.CO↗

A quantitative improvement on the hypergraph Balog-Szemerédi-Gowers theorem

In this note, we obtain a quantitative improvement on the hypergraph variant of the Balog-Szemerédi-Gowers theorem due to Sudakov, Szemerédi, and Vu [Duke Math. J.129.1 (2005): 129--155]. Additionally, we prove the hypergraph variant of the ``almost all'' version of Balog-Szemerédi-Gowers theorem.

math.CO↗

On perfect subdivision tilings

For a given graph $H$, we say that a graph $G$ has a perfect $H$-subdivision tiling if $G$ contains a collection of vertex-disjoint subdivisions of $H$ covering all vertices of $G.$ Let $δ_{\mathrm{sub}}(n, H)$ be the smallest integer $k$ such that any $n$-vertex graph $G$ with minimum degree at least $k$ has a perfect $H$-subdivision tiling. For every graph $H$, we asymptotically determined the value of $δ_{\mathrm{sub}}(n, H)$. More precisely, for every graph $H$ with at least one edge, there is an integer $\mathrm{hcf}_ξ(H)$ and a constant $1 < ξ^*(H)\leq 2$ that can be explicitly determined by structural properties of $H$ such that $δ_{\mathrm{sub}}(n, H) = \left(1 - \frac{1}{ξ^*(H)} + o(1) \right)n$ holds for all $n$ and $H$ unless $\mathrm{hcf}_ξ(H) = 2$ and $n$ is odd. When $\mathrm{hcf}_ξ(H) = 2$ and $n$ is odd, then we show that $δ_{\mathrm{sub}}(n, H) = \left(\frac{1}{2} + o(1) \right)n$.

math.CO↗