Search

The rapid evolution of communication networks in recent decades has intensified the need for advanced Network and Service Management (NSM) strategies to address the growing demands for efficiency, scalability, enhanced performance, and reliability of these networks. Large Language Models (LLMs) have received tremendous attention due to their unparalleled capabilities in various Natural Language Processing (NLP) tasks and generating context-aware insights, offering transformative potential for automating diverse communication NSM tasks. Contrasting existing surveys that consider a single network domain, this survey investigates the integration of LLMs across different communication network domains, including mobile networks and related technologies, vehicular networks, cloud-based networks, and fog/edge-based networks. First, the survey provides foundational knowledge of LLMs, explicitly detailing the generic transformer architecture, general-purpose and domain-specific LLMs, LLM model pre-training and fine-tuning, and their relation to communication NSM. Under a novel taxonomy of network monitoring and reporting, AI-powered network planning, network deployment and distribution, and continuous network support, we extensively categorize LLM applications for NSM tasks in each of the different network domains, exploring existing literature and their contributions thus far. Then, we identify existing challenges and open issues, as well as future research directions for LLM-driven communication NSM, emphasizing the need for scalable, adaptable, and resource-efficient solutions that align with the dynamic landscape of communication networks. We envision that this survey serves as a holistic roadmap, providing critical insights for leveraging LLMs to enhance NSM.

The coordination of vehicles is a crucial element of autonomous driving, as it enhances the efficiency, convenience, and safety of road traffic. In order to fully exploit the capabilities of such coordination, communication with high data rate and low latency is required. It can be reasonably argued that millimeter-wave (mm-wave) vehicle-to-vehicle (V2V) systems are capable of fulfilling the aforementioned requirements. Nevertheless, in order to develop a system that can be deployed in real-world scenarios and to gain an understanding of the various effects of mm-wave propagation, it is necessary to perform radio propagation measurements and to derive radio channel models from them across a range of scenarios and environments. To this end, we have conducted measurement campaigns at 60\,GHz in a variety of situations, including driving in a convoy, driving in opposite direction on a six-lane road, and overtaking. These measurements employ a channel sounder based on ReRoMA, a recently introduced concept that enables the real-time measurement of dynamic double-directional radio channels. The evaluations presented herein encompass key channel parameters, including the path loss (path loss coefficient of approximately 1.9), the root mean square (RMS) delay spread (within a range of 5\,ns to 110\,ns), the angular spreads (in a range of 0.05 to 0.4), the power distribution among multipath components, and the channel stationarity time (multiple seconds)., Comment: 15 pages

Fine-tuning has become a popular approach to adapting large foundational models to specific tasks. As the size of models and datasets grows, parameter-efficient fine-tuning techniques are increasingly important. One of the most widely used methods is Low-Rank Adaptation (LoRA), with adaptation update expressed as the product of two low-rank matrices. While LoRA was shown to possess strong performance in fine-tuning, it often under-performs when compared to full-parameter fine-tuning (FPFT). Although many variants of LoRA have been extensively studied empirically, their theoretical optimization analysis is heavily under-explored. The starting point of our work is a demonstration that LoRA and its two extensions, Asymmetric LoRA and Chain of LoRA, indeed encounter convergence issues. To address these issues, we propose Randomized Asymmetric Chain of LoRA (RAC-LoRA) -- a general optimization framework that rigorously analyzes the convergence rates of LoRA-based methods. Our approach inherits the empirical benefits of LoRA-style heuristics, but introduces several small but important algorithmic modifications which turn it into a provably convergent method. Our framework serves as a bridge between FPFT and low-rank adaptation. We provide provable guarantees of convergence to the same solution as FPFT, along with the rate of convergence. Additionally, we present a convergence analysis for smooth, non-convex loss functions, covering gradient descent, stochastic gradient descent, and federated learning settings. Our theoretical findings are supported by experimental results., Comment: 36 pages, 4 figures, 2 algorithms

In computer-assisted orthopedic surgery (CAOS), accurate pre-operative to intra-operative bone registration is an essential and critical requirement for providing navigational guidance. This registration process is challenging since the intra-operative 3D points are sparse, only partially overlapped with the pre-operative model, and disturbed by noise and outliers. The commonly used method in current state-of-the-art orthopedic robotic system is bony landmarks based registration, but it is very time-consuming for the surgeons. To address these issues, we propose a novel partial-to-full registration framework based on gradient-SDF for CAOS. The simulation experiments using bone models from publicly available datasets and the phantom experiments performed under both optical tracking and electromagnetic tracking systems demonstrate that the proposed method can provide more accurate results than standard benchmarks and be robust to 90% outliers. Importantly, our method achieves convergence in less than 1 second in real scenarios and mean target registration error values as low as 2.198 mm for the entire bone model. Finally, it only requires random acquisition of points for registration by moving a surgical probe over the bone surface without correspondence with any specific bony landmarks, thus showing significant potential clinical value.

The ever-increasing demand for higher data rates in communication systems intensifies the need for advanced non-linear equalizers capable of higher performance. Recently artificial neural networks (ANNs) were introduced as a viable candidate for advanced non-linear equalizers, as they outperform traditional methods. However, they are computationally complex and therefore power hungry. Spiking neural networks (SNNs) started to gain attention as an energy-efficient alternative to ANNs. Recent works proved that they can outperform ANNs at this task. In this work, we explore the design space of an SNN-based decision-feedback equalizer (DFE) to reduce its computational complexity for an efficient implementation on field programmable gate array (FPGA). Our Results prove that it achieves higher communication performance than ANN-based DFE at roughly the same throughput and at 25X higher energy efficiency., Comment: accepted for publication in IEEE Middle East Conference on Communications and Networking (MECOM 2024). November, 2024

The Reggeized pion is expected to provide the main contribution to the forward cross section in light meson photoproduction reactions with charge exchange at high energies. We discuss the Reggeization of pion exchange in charged pion photoproduction with an emphasis on consistency with current conservation. We show that the gauge-invariant amplitude for the exchange of a particle with generic even spin $J\geq 2$ in the $t$-channel is analytic at $J=0$, and that it can be interpreted in terms of the nucleon electric current. This enables us to reconcile the dynamics in the $s$- and $u$-channel, which involves also nucleon exchanges, with the amplitude expressed in terms of $t$-channel partial waves, as required by Regge theory., Comment: 16 pages, 4 figures

Merging Large Language Models (LLMs) is a cost-effective technique for combining multiple expert LLMs into a single versatile model, retaining the expertise of the original ones. However, current approaches often overlook the importance of safety alignment during merging, leading to highly misaligned models. This work investigates the effects of model merging on alignment. We evaluate several popular model merging techniques, demonstrating that existing methods do not only transfer domain expertise but also propagate misalignment. We propose a simple two-step approach to address this problem: (i) generating synthetic safety and domain-specific data, and (ii) incorporating these generated data into the optimization process of existing data-aware model merging techniques. This allows us to treat alignment as a skill that can be maximized in the resulting merged LLM. Our experiments illustrate the effectiveness of integrating alignment-related data during merging, resulting in models that excel in both domain expertise and alignment., Comment: Under review

Forward photoproduction of $\pi^+\pi^-$ pairs with invariant mass of the order of $m_\rho\sim 770$ MeV is traditionally understood to be produced via Pomeron exchange. Based on a detailed analysis of the CLAS photoproduction data, it is shown that the dynamics of two-pion photoproduction for $|t|\gtrsim 0.5$ GeV$^2$ cannot be explained by Pomeron exchange alone. This motivates the development of a new theoretical model of two-pion photoproduction which incorporates both two-pion and pion-nucleon resonant contributions. After fitting free parameters, the model provides an excellent description of the low moments of the angular distribution measured at CLAS, and enables an assessment of the relative contributions of particular production mechanisms and an interpretation of the various features of the data in terms of these mechanisms., Comment: 30 pages, 21 figures

Relieving the reliance of neural network training on a global back-propagation (BP) has emerged as a notable research topic due to the biological implausibility and huge memory consumption caused by BP. Among the existing solutions, local learning optimizes gradient-isolated modules of a neural network with local errors and has been proved to be effective even on large-scale datasets. However, the reconciliation among local errors has never been investigated. In this paper, we first theoretically study non-greedy layer-wise training and show that the convergence cannot be assured when the local gradient in a module w.r.t. its input is not reconciled with the local gradient in the previous module w.r.t. its output. Inspired by the theoretical result, we further propose a local training strategy that successively regularizes the gradient reconciliation between neighboring modules without breaking gradient isolation or introducing any learnable parameters. Our method can be integrated into both local-BP and BP-free settings. In experiments, we achieve significant performance improvements compared to previous methods. Particularly, our method for CNN and Transformer architectures on ImageNet is able to attain a competitive performance with global BP, saving more than 40% memory consumption., Comment: ICML 2024

Dexterous in-hand manipulation is a unique and valuable human skill requiring sophisticated sensorimotor interaction with the environment while respecting stability constraints. Satisfying these constraints with generated motions is essential for a robotic platform to achieve reliable in-hand manipulation skills. Explicitly modelling these constraints can be challenging, but they can be implicitly modelled and learned through experience or human demonstrations. We propose a learning and control approach based on dictionaries of motion primitives generated from human demonstrations. To achieve this, we defined an optimization process that combines motion primitives to generate robot fingertip trajectories for moving an object from an initial to a desired final pose. Based on our experiments, our approach allows a robotic hand to handle objects like humans, adhering to stability constraints without requiring explicit formalization. In other words, the proposed motion primitive dictionaries learn and implicitly embed the constraints crucial to the in-hand manipulation task., Comment: 9 pages, 6 pictures, ROMAN 2024

The design of cell-free massive MIMO (CF-mMIMO) systems requires accurate, measurement-based channel models. This paper provides the first results from the by far most extensive outdoor measurement campaign for CF-mMIMO channels in an urban environment. We measured impulse responses between over 20,000 potential access point (AP) locations and 80 user equipments (UEs) at 3.5 GHz with 350 MHz bandwidth (BW). Measurements use a "virtual array" approach at the AP and a hybrid switched/virtual approach at the UE. This paper describes the sounder design, measurement environment, data processing, and sample results, particularly the evolution of the power-delay profiles (PDPs) as a function of the AP locations, and its relation to the propagation environment., Comment: Submitted to: VTC 2024-Fall

In this paper, we tackle the network delays in the Internet of Things (IoT) for an enhanced QoS through a stable and optimized federated fog computing infrastructure. Network delays contribute to a decline in the Quality-of-Service (QoS) for IoT applications and may even disrupt time-critical functions. Our paper addresses the challenge of establishing fog federations, which are designed to enhance QoS. However, instabilities within these federations can lead to the withdrawal of providers, thereby diminishing federation profitability and expected QoS. Additionally, the techniques used to form federations could potentially pose data leakage risks to end-users whose data is involved in the process. In response, we propose a stable and comprehensive federated fog architecture that considers federated network profiling of the environment to enhance the QoS for IoT applications. This paper introduces a decentralized evolutionary game theoretic algorithm built on top of a Genetic Algorithm mechanism that addresses the fog federation formation issue. Furthermore, we present a decentralized federated learning algorithm that predicts the QoS between fog servers without the need to expose users' location to external entities. Such a predictor module enhances the decision-making process when allocating resources during the federation formation phases without exposing the data privacy of the users/servers. Notably, our approach demonstrates superior stability and improved QoS when compared to other benchmark approaches.

Inclusive production processes will be important for the first observations of $XYZ$ states at new generation electron-hadron colliders, as they generally benefit from larger cross sections than their exclusive counterparts. We make predictions of semi-inclusive photoproduction of the $\chi_{c1}(1P)$ and $X(3872)$, whose peripheral production is assumed to be dominated by vector exchanges. We validate the applicability of Vector Meson Dominance in the axial-vector charmonium sector and calculate production rates at center-of-mass energies relevant for future experimental facilities. We find the semi-inclusive cross sections near threshold to be enhanced by a factor of $\sim 2-3$ compared to the exclusive reaction and well suited for a first observation in photoproduction., Comment: 16 pages, 9 figures. Typos fixed and references updated. Link to code provided. Version appearing in PRD

We explore the impact of training with more diverse datasets, characterized by the number of unique samples, on the performance of self-supervised learning (SSL) under a fixed computational budget. Our findings consistently demonstrate that increasing pretraining data diversity enhances SSL performance, albeit only when the distribution distance to the downstream data is minimal. Notably, even with an exceptionally large pretraining data diversity achieved through methods like web crawling or diffusion-generated data, among other ways, the distribution shift remains a challenge. Our experiments are comprehensive with seven SSL methods using large-scale datasets such as ImageNet and YFCC100M amounting to over 200 GPU days. Code and trained models are available at https://github.com/hammoudhasan/DiversitySSL, Comment: ECCV 2024

Distribution shifts and adversarial examples are two major challenges for deploying machine learning models. While these challenges have been studied individually, their combination is an important topic that remains relatively under-explored. In this work, we study the problem of adversarial robustness under a common setting of distribution shift - unsupervised domain adaptation (UDA). Specifically, given a labeled source domain $D_S$ and an unlabeled target domain $D_T$ with related but different distributions, the goal is to obtain an adversarially robust model for $D_T$. The absence of target domain labels poses a unique challenge, as conventional adversarial robustness defenses cannot be directly applied to $D_T$. To address this challenge, we first establish a generalization bound for the adversarial target loss, which consists of (i) terms related to the loss on the data, and (ii) a measure of worst-case domain divergence. Motivated by this bound, we develop a novel unified defense framework called Divergence Aware adveRsarial Training (DART), which can be used in conjunction with a variety of standard UDA methods; e.g., DANN [Ganin and Lempitsky, 2015]. DART is applicable to general threat models, including the popular $\ell_p$-norm model, and does not require heuristic regularizers or architectural changes. We also release DomainRobust: a testbed for evaluating robustness of UDA models to adversarial attacks. DomainRobust consists of 4 multi-domain benchmark datasets (with 46 source-target pairs) and 7 meta-algorithms with a total of 11 variants. Our large-scale experiments demonstrate that on average, DART significantly enhances model robustness on all benchmarks compared to the state of the art, while maintaining competitive standard accuracy. The relative improvement in robustness from DART reaches up to 29.2% on the source-target domain pairs considered.

We present SynthCLIP, a CLIP model trained on entirely synthetic text-image pairs. Leveraging recent text-to-image (TTI) networks and large language models (LLM), we generate synthetic datasets of images and corresponding captions at scale, with no human intervention. In this work, we provide an analysis on CLIP models trained on synthetic data. We provide insights on the data generation strategy, number of samples required, scaling trends, and resulting properties. We also introduce SynthCI-30M, a purely synthetic dataset comprising 30 million captioned images. Our code, trained models, and data, are released as open source at https://github.com/hammoudhasan/SynthCLIP, Comment: Under review

Data assimilation (DA) plays a pivotal role in diverse applications, ranging from climate predictions and weather forecasts to trajectory planning for autonomous vehicles. A prime example is the widely used ensemble Kalman filter (EnKF), which relies on linear updates to minimize variance among the ensemble of forecast states. Recent advancements have seen the emergence of deep learning approaches in this domain, primarily within a supervised learning framework. However, the adaptability of such models to untrained scenarios remains a challenge. In this study, we introduce a novel DA strategy that utilizes reinforcement learning (RL) to apply state corrections using full or partial observations of the state variables. Our investigation focuses on demonstrating this approach to the chaotic Lorenz '63 system, where the agent's objective is to minimize the root-mean-squared error between the observations and corresponding forecast states. Consequently, the agent develops a correction strategy, enhancing model forecasts based on available system state observations. Our strategy employs a stochastic action policy, enabling a Monte Carlo-based DA framework that relies on randomly sampling the policy to generate an ensemble of assimilated realizations. Results demonstrate that the developed RL algorithm performs favorably when compared to the EnKF. Additionally, we illustrate the agent's capability to assimilate non-Gaussian data, addressing a significant limitation of the EnKF.