As financial systems become increasingly complex and interconnected, traditional fraud detection methods struggle to keep pace with sophisticated fraudulent activities. This paper introduces FraudGNN-RL, an innovative framework that combines Graph Neural Networks (GNNs) with Reinforcement Learning (RL) for adaptive and context-aware financial fraud detection. Our approach models financial transactions as a dynamic graph, where entities (e.g., users, merchants) are nodes and transactions form edges. We propose a novel GNN architecture, Temporal-Spatial-Semantic Graph Convolution (TSSGC), which simultaneously captures temporal patterns, spatial relationships, and semantic information in transaction data. The RL component, implemented as a Deep Q-Network (DQN), dynamically adjusts the fraud detection threshold and feature importance, allowing the model to adapt to evolving fraud patterns and minimize detection costs. We further introduce a Federated Learning scheme to enable collaborative model training across multiple financial institutions while preserving data privacy. Extensive experiments on a large-scale, real-world financial dataset demonstrate that FraudGNN-RL outperforms state-of-the-art baselines, achieving a 97.3% F1-score and reducing false positives by 31% compared to the best-performing baseline. Our framework also shows remarkable resilience to concept drift and adversarial attacks, maintaining high performance over extended periods. These results suggest that FraudGNN-RL offers a robust, adaptive, and privacy-preserving solution for financial fraud detection in the era of Big Data and interconnected financial ecosystems.

This study introduces a novel application of a temporal convolutional network (TCN) for projecting carbon tax prices, addressing the critical need for accurate forecasting in climate policy. Utilizing data from the World Carbon Pricing Database, we demonstrate that the TCN significantly outperformed traditional time series models in capturing the complex dynamics of carbon pricing. Our model achieved a 31.4% improvement in mean absolute error over ARIMA baselines, with an MAE of 2.43 compared to 3.54 for ARIMA. The TCN model also showed superior performance across different time horizons, demonstrating a 30.0% lower MAE for 1-year projections, and enhanced adaptability to policy changes, with only a 39.8% increase in prediction error after major shifts, compared to ARIMA’s 95.6%. These results underscore the potential of deep learning for enhancing the precision of carbon price projections, thereby supporting more informed and effective climate policy decisions. Our findings have significant implications for policymakers and stakeholders in the realm of carbon pricing and climate change mitigation strategies, offering a powerful tool for navigating the complex landscape of environmental economics.

As unmanned aerial vehicle (UAV) technology expands into diverse applications, the demand for enhanced performance intensifies. Blockchain sharding technology offers promising avenues for improving data processing capabilities and security in drone networks. However, the inherent mobility of UAVs and their dynamic operational environment pose significant challenges to conventional sharding techniques, often resulting in communication latencies and data synchronization delays that compromise efficiency. This study presents a novel blockchain-based adaptive sharding framework specifically designed for UAV ecosystems. Our research extends beyond improving data transmission rates to encompass an enhanced Asynchronous Advantage Actor–Critic algorithm, tailored to address long-term optimization objectives in aerial networks. The proposed optimizations focus on dual objectives: enhancing data security while concurrently accelerating processing speeds. By addressing the limitations of traditional approaches, this work aims to facilitate seamless communication and foster innovation in UAV networks. The adaptive sharding framework, coupled with the refined A3C algorithm, presents a comprehensive solution to the unique challenges faced by mobile aerial systems in blockchain implementation.

Lithographic hotspot detection is crucial for ensuring manufacturability and yield in advanced integrated circuit (IC) designs. While machine learning approaches have shown promise, they often struggle with detecting truly-never-seen-before (TNSB) hotspots and reducing false alarms on hard-to-classify (HTC) patterns. This paper presents a novel multi-task deep learning framework for lithographic hotspot detection that addresses these challenges. Our key contributions include: (1) A synthetic pattern generation method based on early design space exploration (EDSE) to augment training data and improve TNSB hotspot detection; (2) A multi-task convolutional neural network architecture that jointly performs hotspot classification and localization; and (3) An adaptive loss function that balances hotspot detection accuracy and false alarm reduction. Experimental results on the ICCAD-2019 benchmark dataset demonstrate that our approach achieves 98.5% accuracy in hotspot detection with only 1.2% false alarm rate, significantly outperforming state-of-the-art methods. Furthermore, we show a 22% improvement in TNSB hotspot detection and a 5X reduction in false alarms on HTC patterns compared to previous techniques. The proposed framework provides a robust solution for lithographic hotspot detection in early stages of IC design, enabling more efficient design-for-manufacturability optimization.