arXiv subjects

Muhammad Shahbaz

Publications and source records attributed to Muhammad Shahbaz.

At least 19 recordsLinked to original sources

arXiv2026

Camera and LiDAR BEV Fusion for Cooperative 3D Object Detection on TUMTraf V2X

We describe a Camera and LiDAR fusion detector developed for the TUMTraf V2X cooperative 3D object detection track of the DriveX 2026 challenge. The detector fuses three roadside cameras with a fused infrastructure-plus-vehicle point cloud in a shared bird's-eye-view space and predicts boxes through a CenterPoint-style head with a generalized IoU regression loss and an IoU quality re-ranking head. Trained on the provided train and validation splits, the model reaches a 3D mAP of 0.85 on the public Codabench test split. While iterating on the system, we observed that 44 of the 50 test frames are also present in the released train (40) and validation (4) splits with their labels. We therefore conducted two additional studies to quantify how this overlap affects the final score: (1) a finetuning run that oversamples the 44 overlapping frames, reaching 0.89 mAP, and (2) a post-processing run that replaces predictions on those frames with the released ground truth, reaching 0.99 mAP (uploaded to our Codabench account for testing but not published on the leaderboard). All three configurations and their per-class results are reported.

cs.CV↗

arXiv2026

Physics-Informed Teacher-Student Ensemble Learning for Traffic State Estimation with a Varying Speed Limit Scenario

Physics-informed deep learning (PIDL) neural networks have shown their capability as a useful instrument for transportation practitioners in utilizing the underlying relationship between the state variables for traffic state estimation (TSE). Another efficient traffic management approach is implementing varying speed limits (VSLs) on transportation corridors to control traffic and mitigate congestion. However, the existing training architecture of PIDL in the literature cannot accommodate the changing traffic characteristics on a freeway with VSL. To tackle this challenge, we propose a novel framework integrating teacher-student ensemble training with PIDL neural networks for TSE under VSL scenarios. The physics of flow conservation law is encoded locally in the teacher models by PIDL, and the student model uses a multi-layer perceptron classifier (MLP) to identify traffic characteristics and selects the ensemble member of PIDL neural networks for TSE. This integrated framework provides a natural solution for capturing the heterogeneity of VSL and accurately addressing the TSE problem. The case study results validate the proposed ensemble approach, demonstrating its superior performance in TSE compared to other popular baseline methods, as indicated by relative L2 error.

cs.LG↗

arXiv2025

OptiNIC: A Resilient and Tail-Optimal RDMA NIC for Distributed ML Workloads

As distributed machine learning (ML) workloads scale to thousands of GPUs connected by high-speed interconnects, tail latency in collective communication has become a major bottleneck. Existing RDMA transports, such as RoCE, IRN, SRNIC, and Falcon, enforce strict reliability and in-order delivery, relying on retransmissions and packet sequencing to ensure correctness. While these approaches work well for general-purpose workloads, they introduce complexity and latency that scale poorly in ML, where even rare packet delays can stall entire model pipelines. We present OptiNIC, a domain-specific RDMA transport that revisits traditional reliability guarantees based on ML's tolerance for partial or missing data. OptiNIC eliminates retransmissions and in-order delivery from the NIC, enabling a best-effort, out-of-order transport model for RDMA. Unlike traditional RDMA, which signals completion only after complete data delivery, OptiNIC introduces adaptive timeouts to trigger forward progress when data may be lost or delayed. OptiNIC retains standard congestion control mechanisms (e.g., DCQCN, EQDS, or Swift) while shifting loss recovery to the ML pipeline itself (e.g., via the Hadamard Transform and Erasure Coding). Our evaluation shows that OptiNIC improves time-to-accuracy (TTA) by 2x and increases throughput by 1.6x for training and inference, respectively, across two public clouds (i.e., Hyperstack and CloudLab). OptiNIC also lowers 99th-percentile latency by 3.5x, cuts BRAM usage by 2.7x, and nearly doubles NIC resilience to faults-delivering a resilient, tail-optimized RDMA transport purpose-built for distributed ML workloads.

cs.DC↗

arXiv2025

Exchange-Correlation Functionals in 2D Materials: Applications, Challenges, and Limitations

The rapid development of two-dimensional (2D) materials has reshaped modern nanoscience, offering properties that differ fundamentally from their bulk counterparts. As experimental discovery accelerates, the need for reliable computational techniques has become increasingly important. Within the framework of density functional theory, this review explores the critical role of exchange-correlation functionals in predicting key material properties such as structural, optoelectronic, magnetic, and thermal. We examine the challenges posed by quantum confinement, anisotropic screening, and van der Waals interactions, which conventional functionals often fail to describe. Advanced approaches, including meta-GGA, hybrid functionals, and many-body perturbation theory (e.g., GW and Bethe-Salpeter equation), are assessed for their improved accuracy in capturing electronic structure and excitonic effects. We further discuss the non-universality of functionals across different 2D material families and the emerging role of machine learning to enhance computational efficiency. Finally, the review outlines current limitations and emerging strategies, providing a roadmap for advancing exchange-correlation functionals and beyond, to enable the practical design and application of 2D materials.

cond-mat.mtrl-sci↗

arXiv2025

Reimagining RDMA Through the Lens of ML

As distributed machine learning (ML) workloads scale to thousands of GPUs connected by ultra-high-speed inter-connects, tail latency in collective communication has emerged as a primary bottleneck. Prior RDMA designs, like RoCE, IRN, and SRNIC, enforce strict reliability and in-order delivery, relying on retransmissions and packet sequencing to ensure correctness. While effective for general-purpose workloads, these mechanisms introduce complexity and latency that scale poorly, where even rare packet losses or delays can consistently degrade system performance. We introduce Celeris, a domain-specific RDMA transport that revisits traditional reliability guarantees based on ML's tolerance for lost or partial data. Celeris removes retransmissions and in-order delivery from the RDMA NIC, enabling best-effort transport that exploits the robustness of ML workloads. It retains congestion control (e.g., DCQCN) and manages communication with software-level mechanisms such as adaptive timeouts and data prioritization, while shifting loss recovery to the ML pipeline (e.g., using the Hadamard Transform). Early results show that Celeris reduces 99th-percentile latency by up to 2.3x, cuts BRAM usage by 67%, and nearly doubles NIC resilience to faults -- delivering a resilient, scalable transport tailored for ML at cluster scale.

cs.DC↗

arXiv2025

TEGRA: A Flexible & Scalable NextGen Mobile Core

To support emerging mobile use cases (e.g., AR/VR, autonomous driving, and massive IoT), next-generation mobile cores for 5G and 6G are being re-architected as service-based architectures (SBAs) running on both private and public clouds. However, current performance optimization strategies for scaling these cores still revert to traditional NFV-based techniques, such as consolidating functions into rigid, monolithic deployments on dedicated servers. This raises a critical question: Is there an inherent tradeoff between flexibility and scalability in an SBA-based mobile core, where improving performance (and resiliency) inevitably comes at the cost of one or the other? To explore this question, we introduce resilient SBA microservices design patterns and state-management strategies, and propose TEGRA -- a high-performance, flexible, and scalable SBA-based mobile core. By leveraging the mobile core's unique position in the end-to-end internet ecosystem (i.e., at the last-mile edge), TEGRA optimizes performance without compromising adaptability. Our evaluation demonstrates that TEGRA achieves significantly lower latencies, processing requests 20x, 11x, and 1.75x faster than traditional SBA core implementations -- free5GC, Open5GS, and Aether, respectively -- all while matching the performance of state-of-the-art cores (e.g., CoreKube) while retaining flexibility. Furthermore, it reduces the complexity of deploying new features, requiring orders of magnitude fewer lines of code (LoCs) compared to existing cores.

cs.NI↗

arXiv2025

UrbanTwin: Synthetic Roadside LiDAR Datasets

This article presents UrbanTwin datasets, high-fidelity, realistic replicas of three public roadside lidar datasets: LUMPI, V2X-Real-IC, and TUMTraf-I. Each UrbanTwin dataset contains 10K annotated frames corresponding to one of the public datasets. Annotations include 3D bounding boxes, instance segmentation labels, and tracking IDs for six object classes, along with semantic segmentation labels for nine classes. These datasets are synthesized using emulated lidar sensors within realistic digital twins, modeled based on surrounding geometry, road alignment at lane level, and the lane topology and vehicle movement patterns at intersections of the actual locations corresponding to each real dataset. Due to the precise digital twin modeling, the synthetic datasets are well aligned with their real counterparts, offering strong standalone and augmentative value for training deep learning models on tasks such as 3D object detection, tracking, and semantic and instance segmentation. We evaluate the alignment of the synthetic replicas through statistical and structural similarity analysis with real data, and further demonstrate their utility by training 3D object detection models solely on synthetic data and testing them on real, unseen data. The high similarity scores and improved detection performance, compared to models trained on real data, indicate that the UrbanTwin datasets effectively enhance existing benchmark datasets by increasing sample size and scene diversity. In addition, the digital twins can be adapted to test custom scenarios by modifying the design and dynamics of the simulations. To our knowledge, these are the first digitally synthesized datasets that can replace in-domain real-world datasets for lidar perception tasks. UrbanTwin datasets are publicly available at https://dataverse.harvard.edu/dataverse/ucf-ut.

cs.CV↗

arXiv2025

LiGuard: A Streamlined Open-Source Framework for Rapid & Interactive Lidar Research

There is a growing interest in the development of lidar-based autonomous mobility and Intelligent Transportation Systems (ITS). To operate and research on lidar data, researchers often develop code specific to application niche. This approach leads to duplication of efforts across studies that, in many cases, share multiple methodological steps such as data input/output (I/O), pre/post processing, and common algorithms in multi-stage solutions. Moreover, slight changes in data, algorithms, and/or research focus may force major revisions in the code. To address these challenges, we present LiGuard, an open-source software framework that allows researchers to: 1) rapidly develop code for their lidar-based projects by providing built-in support for data I/O, pre/post processing, and commonly used algorithms, 2) interactively add/remove/reorder custom algorithms and adjust their parameters, and 3) visualize results for classification, detection, segmentation, and tracking tasks. Moreover, because it creates all the code files in structured directories, it allows easy sharing of entire projects or even the individual components to be reused by other researchers. The effectiveness of LiGuard is demonstrated via case studies.

cs.CV↗

arXiv2025

UrbanTwin: Building High-Fidelity Digital Twins for Sim2Real LiDAR Perception and Evaluation

LiDAR-based perception in intelligent transportation systems (ITS) relies on deep neural networks trained with large-scale labeled datasets. However, creating such datasets is expensive, time-consuming, and labor-intensive, limiting the scalability of perception systems. Sim2Real learning offers a scalable alternative, but its success depends on the simulation's fidelity to real-world environments, dynamics, and sensors. This tutorial introduces a reproducible workflow for building high-fidelity digital twins (HiFi DTs) to generate realistic synthetic datasets. We outline practical steps for modeling static geometry, road infrastructure, and dynamic traffic using open-source resources such as satellite imagery, OpenStreetMap, and sensor specifications. The resulting environments support scalable and cost-effective data generation for robust Sim2Real learning. Using this workflow, we have released three synthetic LiDAR datasets, namely UT-LUMPI, UT-V2X-Real, and UT-TUMTraf-I, which closely replicate real locations and outperform real-data-trained baselines in perception tasks. This guide enables broader adoption of HiFi DTs in ITS research and deployment.

cs.CV↗

arXiv2025

High-Fidelity Digital Twins for Bridging the Sim2Real Gap in LiDAR-Based ITS Perception

Sim2Real domain transfer offers a cost-effective and scalable approach for developing LiDAR-based perception (e.g., object detection, tracking, segmentation) in Intelligent Transportation Systems (ITS). However, perception models trained in simulation often under perform on real-world data due to distributional shifts. To address this Sim2Real gap, this paper proposes a high-fidelity digital twin (HiFi DT) framework that incorporates real-world background geometry, lane-level road topology, and sensor-specific specifications and placement. We formalize the domain adaptation challenge underlying Sim2Real learning and present a systematic method for constructing simulation environments that yield in-domain synthetic data. An off-the-shelf 3D object detector is trained on HiFi DT-generated synthetic data and evaluated on real data. Our experiments show that the DT-trained model outperforms the equivalent model trained on real data by 4.8%. To understand this gain, we quantify distributional alignment between synthetic and real data using multiple metrics, including Chamfer Distance (CD), Maximum Mean Discrepancy (MMD), Earth Mover's Distance (EMD), and Fr'echet Distance (FD), at both raw-input and latent-feature levels. Results demonstrate that HiFi DTs substantially reduce domain shift and improve generalization across diverse evaluation scenarios. These findings underscore the significant role of digital twins in enabling reliable, simulation-based LiDAR perception for real-world ITS applications.

cs.CV↗

arXiv2025

SpliDT: Partitioned Decision Trees for Scalable Stateful Inference at Line Rate

Machine learning (ML) is increasingly being deployed in programmable data planes (switches and SmartNICs) to enable real-time traffic analysis, security monitoring, and in-network decision-making. Decision trees (DTs) are particularly well-suited for these tasks due to their interpretability and compatibility with data-plane architectures, i.e., match-action tables (MATs). However, existing in-network DT implementations are constrained by the need to compute all input features upfront, forcing models to rely on a small, fixed set of features per flow. This significantly limits model accuracy and scalability under stringent hardware resource constraints. We present SPLIDT, a system that rethinks DT deployment in the data plane by enabling partitioned inference over sliding windows of packets. SPLIDT introduces two key innovations: (1) it assigns distinct, variable feature sets to individual sub-trees of a DT, grouped into partitions, and (2) it leverages an in-band control channel (via recirculation) to reuse data-plane resources (both stateful registers and match keys) across partitions at line rate. These insights allow SPLIDT to scale the number of stateful features a model can use without exceeding hardware limits. To support this architecture, SPLIDT incorporates a custom training and design-space exploration (DSE) framework that jointly optimizes feature allocation, tree partitioning, and DT model depth. Evaluation across multiple real-world datasets shows that SPLIDT achieves higher accuracy while supporting up to 5x more stateful features than prior approaches (e.g., NetBeacon and Leo). It maintains the same low time-to-detection (TTD) as these systems, while scaling to millions of flows with minimal recirculation overhead (<0.05%).

cs.NI↗

arXiv2025

Attention-Based Ensemble Learning for Crop Classification Using Landsat 8-9 Fusion

Remote sensing offers a highly effective method for obtaining accurate information on total cropped area and crop types. The study focuses on crop cover identification for irrigated regions of Central Punjab. Data collection was executed in two stages: the first involved identifying and geocoding six target crops through field surveys conducted in January and February 2023. The second stage involved acquiring Landsat 8-9 imagery for each geocoded field to construct a labelled dataset. The satellite imagery underwent extensive pre-processing, including radiometric calibration for reflectance values, atmospheric correction, and georeferencing verification to ensure consistency within a common coordinate system. Subsequently, image fusion techniques were applied to combine Landsat 8 and 9 spectral bands, creating a composite image with enhanced spectral information, followed by contrast enhancement. During data acquisition, farmers were interviewed, and fields were meticulously mapped using GPS instruments, resulting in a comprehensive dataset of 50,835 data points. This dataset facilitated the extraction of vegetation indices such as NDVI, SAVO, RECI, and NDRE. These indices and raw reflectance values were utilized for classification modeling using conventional classifiers, ensemble learning, and artificial neural networks. A feature selection approach was also incorporated to identify the optimal feature set for classification learning. This study demonstrates the effectiveness of combining remote sensing data and advanced modeling techniques to improve crop classification accuracy in irrigated agricultural regions.

cs.CV↗

arXiv2025

DINOSTAR: Deep Iterative Neural Object Detector Self-Supervised Training for Roadside LiDAR Applications

Recent advancements in deep-learning methods for object detection in point-cloud data have enabled numerous roadside applications, fostering improvements in transportation safety and management. However, the intricate nature of point-cloud data poses significant challenges for human-supervised labeling, resulting in substantial expenditures of time and capital. This paper addresses the issue by developing an end-to-end, scalable, and self-supervised framework for training deep object detectors tailored for roadside point-cloud data. The proposed framework leverages self-supervised, statistically modeled teachers to train off-the-shelf deep object detectors, thus circumventing the need for human supervision. The teacher models follow fine-tuned set standard practices of background filtering, object clustering, bounding-box fitting, and classification to generate noisy labels. It is presented that by training the student model over the combined noisy annotations from multitude of teachers enhances its capacity to discern background/foreground more effectively and forces it to learn diverse point-cloud-representations for object categories of interest. The evaluations, involving publicly available roadside datasets and state-of-art deep object detectors, demonstrate that the proposed framework achieves comparable performance to deep object detectors trained on human-annotated labels, despite not utilizing such human-annotations in its training process.

cs.CV↗

arXiv2023

OptiReduce: Resilient and Tail-Optimal AllReduce for Distributed Deep Learning in the Cloud

We present OptiReduce, a new collective-communication system for the cloud with bounded, predictable completion times for deep-learning jobs in the presence of varying computation (stragglers) and communication (congestion and gradient drops) variabilities. OptiReduce exploits the inherent resiliency and the stochastic nature of distributed deep-learning (DDL) training and fine-tuning to work with approximated (or lost) gradients -- providing an efficient balance between (tail) performance and the resulting accuracy of the trained models. Exploiting this domain-specific characteristic of DDL, OptiReduce introduces (1) mechanisms (e.g., unreliable bounded transport with adaptive timeout) to improve the DDL jobs' tail execution time, and (2) strategies (e.g., Transpose AllReduce and Hadamard Transform) to mitigate the impact of gradient drops on model accuracy. Our evaluation shows that OptiReduce achieves 70% and 30% faster time-to-accuracy (TTA), on average, when operating in shared, cloud environments (e.g., CloudLab) compared to Gloo and NCCL, respectively.

cs.DC↗

arXiv2022

Characterizing and Modeling Control-Plane Traffic for Mobile Core Network

In this paper, we first carry out to our knowledge the first in-depth characterization of control-plane traffic, using a real-world control-plane trace for 37,325 UEs sampled at a real-world LTE Mobile Core Network (MCN). Our analysis shows that control events exhibit significant diversity in device types and time-of-day among UEs. Second, we study whether traditional probability distributions that have been widely adopted for modeling Internet traffic can model the control-plane traffic originated from individual UEs. Our analysis shows that the inter-arrival time of the control events as well as the sojourn time in the UE states of EMM and ECM for the cellular network cannot be modeled as Poisson processes or other traditional probability distributions. We further show that the reasons that these models fail to capture the control-plane traffic are due to its higher burstiness and longer tails in the cumulative distribution than the traditional models. Third, we propose a two-level hierarchical state-machine-based traffic model for UE clusters derived from our adaptive clustering scheme based on the Semi-Markov Model to capture key characteristics of mobile network control-plane traffic -- in particular, the dependence among events generated by each UE, and the diversity in device types and time-of-day among UEs. Finally, we show how our model can be easily adjusted from LTE to 5G to support modeling 5G control-plane traffic, when the sizable control-plane trace for 5G UEs becomes available to train the adjusted model. The developed control-plane traffic generator for LTE/5G networks is open-sourced to the research community to support high-performance MCN architecture design R&D.

cs.NI↗

arXiv2022

Enabling the Reflex Plane with the nanoPU

Many recent papers have demonstrated fast in-network computation using programmable switches, running many orders of magnitude faster than CPUs. The main limitation of writing software for switches is the constrained programming model and limited state. In this paper we explore whether a new type of CPU, called the nanoPU, offers a useful middle ground, with a familiar C/C++ programming model, and potentially many terabits/second of packet processing on a single chip, with an RPC response time less than 1 $\mu$s. To evaluate the nanoPU, we prototype and benchmark three common network services: packet classification, network telemetry report processing, and consensus protocols on the nanoPU. Each service is evaluated using cycle-accurate simulations on FPGAs in AWS. We found that packets are classified 2$\times$ faster and INT reports are processed more than an order of magnitude quickly than state-of-the-art approaches. Our production quality Raft consensus protocol, running on the nanoPU, writes to a 3-way replicated key-value store (MICA) in 3 $\mu$s, twice as fast as the state-of-the-art, with 99\% tail latency of only 3.26 $\mu$s. To understand how these services can be combined, we study the design and performance of a {\em network reflex plane}, designed to process telemetry data, make fast control decisions, and update consistent, replicated state within a few microseconds.

cs.NI↗

arXiv2022

Homunculus: Auto-Generating Efficient Data-Plane ML Pipelines for Datacenter Networks

Support for Machine Learning (ML) applications in networks has significantly improved over the last decade. The availability of public datasets and programmable switching fabrics (including low-level languages to program them) present a full-stack to the programmer for deploying in-network ML. However, the diversity of tools involved, coupled with complex optimization tasks of ML model design and hyperparameter tuning while complying with the network constraints (like throughput and latency), put the onus on the network operator to be an expert in ML, network design, and programmable hardware. This multi-faceted nature of in-network tools and expertise in ML and hardware is a roadblock for ML to become mainstream in networks, today. We present Homunculus, a high-level framework that enables network operators to specify their ML requirements in a declarative, rather than imperative way. Homunculus takes as input, the training data and accompanying network constraints, and automatically generates and installs a suitable model onto the underlying switching hardware. It performs model design-space exploration, training, and platform code-generation as compiler stages, leaving network operators to focus on acquiring high-quality network data. Our evaluations on real-world ML applications show that Homunculus's generated models achieve up to 12% better F1 score compared to hand-tuned alternatives, while requiring only 30 lines of single-script code on average. We further demonstrate the performance of the generated models on emerging per-packet ML platforms to showcase its timely and practical significance.

cs.NI↗

arXiv2020

The nanoPU: Redesigning the CPU-Network Interface to Minimize RPC Tail Latency

The nanoPU is a new networking-optimized CPU designed to minimize tail latency for RPCs. By bypassing the cache and memory hierarchy, the nanoPU directly places arriving messages into the CPU register file. The wire-to-wire latency through the application is just 65ns, about 13x faster than the current state-of-the-art. The nanoPU moves key functions from software to hardware: reliable network transport, congestion control, core selection, and thread scheduling. It also supports a unique feature to bound the tail latency experienced by high-priority applications. Our prototype nanoPU is based on a modified RISC-V CPU; we evaluate its performance using cycle-accurate simulations of 324 cores on AWS FPGAs, including real applications (MICA and chain replication).

cs.AR↗