arXiv subjects

Xiaoyi Wu

Publications and source records attributed to Xiaoyi Wu.

13 recordsLinked to original sources

arXiv2026

Hybrid Feedback-Guided Optimal Learning for Wireless Interactive Panoramic Scene Delivery

Immersive applications such as virtual and augmented reality impose stringent requirements on frame rate, latency, and synchronization between physical and virtual environments. To meet these requirements, an edge server must render panoramic content, predict user head motion, and transmit a portion of the scene that is large enough to cover the user viewport while remaining within wireless bandwidth constraints. Each portion produces two feedback signals: prediction feedback, indicating whether the selected portion covers the actual viewport, and transmission feedback, indicating whether the corresponding packets are successfully delivered. Prior work models this problem as a multi-armed bandit with two-level bandit feedback, but fails to exploit the fact that prediction feedback can be retrospectively computed for all candidate portions once the user head pose is observed. As a result, prediction feedback constitutes full-information feedback rather than bandit feedback. Motivated by this observation, we introduce a two-level hybrid feedback model that combines full-information and bandit feedback, and formulate the portion selection problem as an online learning task under this setting. We derive an instance-dependent regret lower bound for the hybrid feedback model and propose AdaPort, a hybrid learning algorithm that leverages both feedback types to improve learning efficiency. We further establish an instance-dependent regret upper bound that matches the lower bound asymptotically, and demonstrate through measurements on an end-to-end testbed that AdaPort outperforms state-of-the-art learning-based baselines as well as the heuristic minimum scene delivery scheme.

cs.LG↗

arXiv2026

Efficient User Association and Wireless Scheduling with Shorter Time-Scale Rate Adaptation

Rate adaptation is a crucial mechanism in IEEE 802.11 networks and next-generation cellular systems. Since the time scale for rate adaptation is typically much shorter than that for user association and scheduling, we investigate a joint design of wireless user association and scheduling and rate adaptation across different time scales to maximize cumulative network throughput while ensuring desired fairness among users. We develop a MaxWeight-type user association and scheduling algorithm that integrates virtual queues -- tracking each user's scheduling debt to maintain fairness -- and Upper Confidence Bound (UCB) estimates in its weight measure. Each selected user then employs the UCB algorithm for rate adaptation on a short time scale. Our theoretical findings reveal that the proposed algorithm achieves cumulative regret that grows with the square root of the time horizon up to a logarithmic factor and results in zero cumulative fairness violation after a certain number of time frames. Furthermore, since the MaxWeight-type algorithm involves evaluating all the feasible schedules that can be exponential to the number of users due to the interference constraints, leading to high computational complexity, we introduce a low-complexity alternative utilizing the so-called pick-and-compare (PC) approach. We demonstrate the effectiveness of both algorithms through simulations based on real-world data traces.

cs.NI↗

arXiv2026

Exclusive $J/ψ$ photoproduction in photon-proton diffractive scattering: A light-front Hamiltonian approach

We investigate the cross-section for exclusive $J/ψ$ production in photon-proton diffractive scattering within the Basis Light-Front Quantization (BLFQ) framework. The leading-order contribution to this process is well approximated by the charge conjugation-even two-gluon (``pomeron") exchange mechanism in the dipole model, which factorizes the total amplitude into the dipole scattering amplitude and the convolution of the $J/ψ$ and photon light-front wave functions (LFWFs). We express the dipole scattering amplitude as the matrix element of gluon field operators inserted between proton states, with the element being sensitive to the proton LFWFs and the Bjorken scaling variable, $x$. The proton and $J/ψ$ LFWFs are obtained by diagonalizing their respective light-front Hamiltonians within the BLFQ approach, while the virtual photon LFWFs are employed from perturbative QCD. Our results provide initial conditions for the Balitsky-Kovchegov (BK) equation, which can be used to probe the proton structure at smaller Bjorken scales. This work offers valuable theoretical insights for future electron-ion collider experiments.

hep-ph↗

arXiv2026

Nuclear matter and proton parton distributions in a light-front Hamiltonian framework

We develop a light-front Hamiltonian formulation of symmetric nuclear matter within the quark-meson coupling model, using Basis Light-Front Quantization to solve the in-medium nucleon eigenvalue problem. The Hamiltonian incorporates confinement in the valence sector and is truncated to include up to one dynamical gluon. Medium effects are introduced via scalar and vector mean fields, yielding a self-consistent, density-dependent effective quark mass and modified nucleon structure. The resulting energy per nucleon, pressure, and incompressibility are consistent with empirical constraints at the saturation point. At nuclear saturation density, the gluon probability in the nucleon wave function increases slightly, while the valence probability and quark momentum fraction decrease. The unpolarized quark and gluon distributions show a noticeable enhancement at large momentum fraction ($x \gtrsim 0.4$), illustrated at an evolved scale of $Q^{2} = 10 \mathrm{GeV}^{2}$.

hep-ph↗

arXiv2025

On the Regularity and Fairness of Combinatorial Multi-Armed Bandit

The combinatorial multi-armed bandit model is designed to maximize cumulative rewards in the presence of uncertainty by activating a subset of arms in each round. This paper is inspired by two critical applications in wireless networks, where it's not only essential to maximize cumulative rewards but also to guarantee fairness among arms (i.e., the minimum average reward required by each arm) and ensure reward regularity (i.e., how often each arm receives the reward). In this paper, we propose a parameterized regular and fair learning algorithm to achieve these three objectives. In particular, the proposed algorithm linearly combines virtual queue-lengths (tracking the fairness violations), Time-Since-Last-Reward (TSLR) metrics, and Upper Confidence Bound (UCB) estimates in its weight measure. Here, TSLR is similar to age-of-information and measures the elapsed number of rounds since the last time an arm received a reward, capturing the reward regularity performance, and UCB estimates are utilized to balance the tradeoff between exploration and exploitation in online learning. By exploring a key relationship between virtual queue-lengths and TSLR metrics and utilizing several non-trivial Lyapunov functions, we analytically characterize zero cumulative fairness violation, reward regularity, and cumulative regret performance under our proposed algorithm. These theoretical outcomes are verified by simulations based on two real-world datasets.

cs.LG↗

arXiv2025

EFEAR-4D: Ego-Velocity Filtering for Efficient and Accurate 4D radar Odometry

Odometry is a crucial component for successfully implementing autonomous navigation, relying on sensors such as cameras, LiDARs and IMUs. However, these sensors may encounter challenges in extreme weather conditions, such as snowfall and fog. The emergence of FMCW radar technology offers the potential for robust perception in adverse conditions. As the latest generation of FWCW radars, the 4D mmWave radar provides point cloud with range, azimuth, elevation, and Doppler velocity information, despite inherent sparsity and noises in the point cloud. In this paper, we propose EFEAR-4D, an accurate, highly efficient, and learning-free method for large-scale 4D radar odometry estimation. EFEAR-4D exploits Doppler velocity information delicately for robust ego-velocity estimation, resulting in a highly accurate prior guess. EFEAR-4D maintains robustness against point-cloud sparsity and noises across diverse environments through dynamic object removal and effective region-wise feature extraction. Extensive experiments on two publicly available 4D radar datasets demonstrate state-of-the-art reliability and localization accuracy of EFEAR-4D under various conditions. Furthermore, we have collected a dataset following the same route but varying installation heights of the 4D radar, emphasizing the significant impact of radar height on point cloud quality - a crucial consideration for real-world deployments. Our algorithm and dataset will be available soon at https://github.com/CLASS-Lab/EFEAR-4D.

cs.RO↗

arXiv2025

4D-ROLLS: 4D Radar Occupancy Learning via LiDAR Supervision

A comprehensive understanding of 3D scenes is essential for autonomous vehicles (AVs), and among various perception tasks, occupancy estimation plays a central role by providing a general representation of drivable and occupied space. However, most existing occupancy estimation methods rely on LiDAR or cameras, which perform poorly in degraded environments such as smoke, rain, snow, and fog. In this paper, we propose 4D-ROLLS, the first weakly supervised occupancy estimation method for 4D radar using the LiDAR point cloud as the supervisory signal. Specifically, we introduce a method for generating pseudo-LiDAR labels, including occupancy queries and LiDAR height maps, as multi-stage supervision to train the 4D radar occupancy estimation model. Then the model is aligned with the occupancy map produced by LiDAR, fine-tuning its accuracy in occupancy estimation. Extensive comparative experiments validate the exceptional performance of 4D-ROLLS. Its robustness in degraded environments and effectiveness in cross-dataset training are qualitatively demonstrated. The model is also seamlessly transferred to downstream tasks BEV segmentation and point cloud occupancy prediction, highlighting its potential for broader applications. The lightweight network enables 4D-ROLLS model to achieve fast inference speeds at about 30 Hz on a 4060 GPU. The code of 4D-ROLLS will be made available at https://github.com/CLASS-Lab/4D-ROLLS.

cs.CV↗

arXiv2025

Deep Reinforcement Learning for Day-to-day Dynamic Tolling in Tradable Credit Schemes

Tradable credit schemes (TCS) are an increasingly studied alternative to congestion pricing, given their revenue neutrality and ability to address issues of equity through the initial credit allocation. Modeling TCS to aid future design and implementation is associated with challenges involving user and market behaviors, demand-supply dynamics, and control mechanisms. In this paper, we focus on the latter and address the day-to-day dynamic tolling problem under TCS, which is formulated as a discrete-time Markov Decision Process and solved using reinforcement learning (RL) algorithms. Our results indicate that RL algorithms achieve travel times and social welfare comparable to the Bayesian optimization benchmark, with generalization across varying capacities and demand levels. We further assess the robustness of RL under different hyperparameters and apply regularization techniques to mitigate action oscillation, which generates practical tolling strategies that are transferable under day-to-day demand and supply variability. Finally, we discuss potential challenges such as scaling to large networks, and show how transfer learning can be leveraged to improve computational efficiency and facilitate the practical deployment of RL-based TCS solutions.

cs.LG↗

arXiv2025

LLMER: Crafting Interactive Extended Reality Worlds with JSON Data Generated by Large Language Models

The integration of Large Language Models (LLMs) like GPT-4 with Extended Reality (XR) technologies offers the potential to build truly immersive XR environments that interact with human users through natural language, e.g., generating and animating 3D scenes from audio inputs. However, the complexity of XR environments makes it difficult to accurately extract relevant contextual data and scene/object parameters from an overwhelming volume of XR artifacts. It leads to not only increased costs with pay-per-use models, but also elevated levels of generation errors. Moreover, existing approaches focusing on coding script generation are often prone to generation errors, resulting in flawed or invalid scripts, application crashes, and ultimately a degraded user experience. To overcome these challenges, we introduce LLMER, a novel framework that creates interactive XR worlds using JSON data generated by LLMs. Unlike prior approaches focusing on coding script generation, LLMER translates natural language inputs into JSON data, significantly reducing the likelihood of application crashes and processing latency. It employs a multi-stage strategy to supply only the essential contextual information adapted to the user's request and features multiple modules designed for various XR tasks. Our preliminary user study reveals the effectiveness of the proposed system, with over 80% reduction in consumed tokens and around 60% reduction in task completion time compared to state-of-the-art approaches. The analysis of users' feedback also illuminates a series of directions for further optimization.

cs.MM↗

arXiv2025

On the Low-Complexity of Fair Learning for Combinatorial Multi-Armed Bandit

Combinatorial Multi-Armed Bandit with fairness constraints is a framework where multiple arms form a super arm and can be pulled in each round under uncertainty to maximize cumulative rewards while ensuring the minimum average reward required by each arm. The existing pessimistic-optimistic algorithm linearly combines virtual queue-lengths (tracking the fairness violations) and Upper Confidence Bound estimates as a weight for each arm and selects a super arm with the maximum total weight. The number of super arms could be exponential to the number of arms in many scenarios. In wireless networks, interference constraints can cause the number of super arms to grow exponentially with the number of arms. Evaluating all the feasible super arms to find the one with the maximum total weight can incur extremely high computational complexity in the pessimistic-optimistic algorithm. To avoid this, we develop a low-complexity fair learning algorithm based on the so-called pick-and-compare approach that involves randomly picking $M$ feasible super arms to evaluate. By setting $M$ to a constant, the number of comparison steps in the pessimistic-optimistic algorithm can be reduced to a constant, thereby significantly reducing the computational complexity. Our theoretical proof shows this low-complexity design incurs only a slight sacrifice in fairness and regret performance. Finally, we validate the theoretical result by extensive simulations.

cs.LG↗

arXiv2024

Pion to photon transition form factor: Beyond valence quarks

We investigate the singly virtual transition form factor (TFF) for the $π^0\toγ^*γ$ process in the space-like region using the hard-scattering formalism within the Basis Light-Front Quantization (BLFQ) framework. This form factor is expressed in terms of the perturbatively calculable hard-scattering amplitudes (HSAs) and the light-front wave functions (LFWFs) of the pion. We obtain the pion LFWFs by diagonalizing the light-front QCD Hamiltonian, which is determined for its constituent quark-antiquark and quark-antiquark-gluon Fock sectors with a three-dimensional confinement. We employ the HSAs up to next-to-leading order (NLO) in the quark-antiquark Fock sector and leading order (LO) in the quark-antiquark-gluon Fock sector. The NLO correction to the TFF in the quark-antiquark Fock sector is of the same order as the LO contribution to the TFF in the quark-antiquark-gluon Fock sector. We find that while the quark-antiquark-gluon Fock sector has minimal effect in the large momentum transfer ($Q^2$) region, it has a noteworthy impact in the low-$Q^2$ region. Our results show that, after accounting for both Fock sectors, the TFF within the BLFQ framework aligns well with existing experimental data, particularly in the low $Q^2$ region.

hep-ph↗

arXiv2024

Joint Design of Conventional Public Transport Network and Mobility on Demand

Conventional Public Transport (PT) is based on fixed lines, running with routes and schedules determined a-priori. In low-demand areas, conventional PT is inefficient. Therein, Mobility on Demand (MoD) could serve users more efficiently and with an improved quality of service (QoS). The idea of integrating MoD into PT is therefore abundantly discussed by researchers and practitioners, mainly in the form of adding MoD on top of PT. Efficiency can be instead gained if also conventional PT lines are redesigned after integrating MoD in the first or last mile. In this paper we focus on this re-design problem. We devise a bilevel optimization problem where, given a certain initial design, the upper level determines stop selection and frequency settings, while the lower level routes a fleet of MoD vehicles. We propose a solution method based on Particle Swarm Optimization (PSO) for the upper level, while we adopt Large Neighborhood Search (LNS) in the lower level. Our solution method is computationally efficient and we test it in simulations with up to 10k travel requests. Results show important operational cost savings obtained via appropriately reducing the conventional PT coverage after integrating MoD, while preserving QoS.

eess.SY↗

arXiv2023

RH-Map: Online Map Construction Framework of Dynamic Objects Removal Based on Region-wise Hash Map Structure

Mobile robots navigating in outdoor environments frequently encounter the issue of undesired traces left by dynamic objects and manifested as obstacles on map, impeding robots from achieving accurate localization and effective navigation. To tackle the problem, a novel map construction framework based on 3D region-wise hash map structure (RH-Map) is proposed, consisting of front-end scan fresher and back-end removal modules, which realizes real-time map construction and online dynamic object removal (DOR). First, a two-layer 3D region-wise hash map structure of map management is proposed for effective online DOR. Then, in scan fresher, region-wise ground plane estimation (R-GPE) is adopted for estimating and preserving ground information and Scan-to-Map Removal (S2M-R) is proposed to discriminate and remove dynamic regions. Moreover, the lightweight back-end removal module maintaining keyframes is proposed for further DOR. As experimentally verified on SemanticKITTI, our proposed framework yields promising performance on online DOR of map construction compared with the state-of-the-art methods. And we also validate the proposed framework in real-world environments.

cs.RO↗