edge computing

UOIM: Online Incentive Design for Fair Hierarchical Federated Learning and Unlearning in Edge Computing

Hierarchical Federated Learning (HFL) has emerged as a promising evolution of Federated Learning (FL) where a shared model is learned collaboratively via hierarchical aggregation, and it requires federated unlearning to protect users’ right to be forgotten in the training process. However, most existing designs neglect...

Generative AI Model Assisted Multimodal Semantic Vehicular Edge Computing for Collaborative Perception

Vehicular edge computing (VEC) plays a pivotal role in enabling cooperative perception for connected autonomous vehicles (CAVs), providing comprehensive environmental awareness for safe vehicle control and road safety. However, collaborative perception in VEC faces significant challenges arising from stringent bandwidt...

Scalable persistence pairing graphs on scalar fields with an application to porous media

We present a scalable method for computing persistence pairing graphs from large cubical filtrations and use it to test whether graph organization among persistent classes contains information beyond persistence intervals alone. The method cancels equal-value pairs before global reduction, constructs the compressed com...

Settling PROPm and PROPavg in Graphical Resource Allocation

We study proportional fairness in graphical resource allocation, where agents are vertices, indivisible items are edges, and each item must be allocated to one of its two endpoints. It has been proved that PROP1 orientations always exist and PROPx orientations may not, but it has remained open whether the intermediate...

Breaking the $\sqrt{3}$ Barrier for Maximum Weighted $3$-Set Packing

A deterministic polynomial-time approximation for Maximum Weighted $3-Set Packing, breaking the $\sqrt3$ locality-gap barrier of squared-weight local search and proving that $\sqrt3$ is a locality-gap lower bound for the squared-weight objective even with exchanges of arbitrary size.

Deadline-Aware Multi-Agent Reinforcement Learning for TSN-Based Vehicular Edge Networks

This work proposes a multi-agent reinforcement learning (MARL) approach for queue-level scheduling in TSN-enabled VEC and employs multi-agent proximal policy optimization (MAPPO) to enable coordinated yet autonomous scheduling decisions.

Heteroepitaxy of 2D MoO2 single crystals on A-plane sapphires for In-Sensor Edge Computing

Two-dimensional (2D) non-layered oxides, such as MoO2, are emerging candidates for next-generation electronics, optoelectronics and sensors, especially novel in-sensor computing. However, there is a lack of effective method to grow large-area 2D MoO2 single crystals due to multiple preferred orientations induced by unc...

MODEL ADOPTION #3822: REJECT — LEDGER BENCH: Apply filter: Pool density HIGH (>=90) — improves win rate from 32.5% to 34. — E8 Intelligence Research

DECISION: REJECT ELEMENT: No element survives — the pool-density ≥90 filter is a statistical artifact, not an E8-geometry-compatible edge. EXPECTED IMPACT: The +1.5pp backtest gain (n=1800) is within noise for a 32.5%→34.0% shift (σ≈1.1pp). Merlin's own live-window projection shows only +8.6pp on 7 trades (n=7, meaning...

YOLOv11-GPC: A Lightweight Detection Model for Maize Diseases and Pests with Edge-Device Deployment on Jetson Orin Nano

This study designed an edge computing detection system for maize pests and diseases based on an improved YOLOv11 model and designed a C2PSLCA module based on linear self-attention to replace the C2PSA module in YOLOv11, enhancing the model’s ability to extract features of maize pests and diseases in complex environment...

Machine Learning and IOT-Driven Decision Support Systems in Agriculture: A Systematic Review of Tabular Telemetry and Predictive Modeling

Decision Support Systems (DSS) in precision agriculture have evolved from static rule-based engines into data-driven frameworks powered by Internet of Things (IoT) telemetry and supervised machine learning. While recent literature increasingly focuses on deep learning and volatile commodity forecasting, empirical farm-...

Deliverable No D3.5 Production Planning & Reconfiguration v1

Deliverable D3.5 Production Planning & Reconfiguration v1 provides an overall status update of two core components developed within the CIRCMAN5.0 project’s Work Package 3 (WP3). They are aligned with WP3’s purpose that is to deliver novel simulation and modelling software to improve process and product manufacturing....

Physical neural networks with trainable nonlinear connections for continuous control