Journal of Intelligence and Engineering Systems

Optimizing Neural Network Architectures for Autonomous Systems

2026-02-26T18:09:32+02:00

The research focuses on optimizing neural network architectures to improve the decision-making capabilities of autonomous systems. Through a series of simulations and real-world tests, we identify key architectural modifications that enhance efficiency and accuracy. The findings indicate that specific architecture designs can significantly impact system performance, leading to more reliable and intelligent autonomous applications. This study provides a framework for future research in neural adaptive systems.

This is a free preview. The complete article is available with a valid subscription.

Hierarchical Neuro-Symbolic Constraint Propagation Framework for Real-Time Autonomous Decision Engineering in Multi-Agent Cyber-Physical Systems

2026-06-26T17:41:28+03:00

The integration of neuro-symbolic reasoning with constraint propagation in multi-agent cyber-physical systems (CPS) presents critical scalability and inference latency challenges unresolved by existing architectures. This paper proposes a Hierarchical Neuro-Symbolic Constraint Propagation Framework (HNS-CPF) that couples deep graph neural networks with symbolic Satisfiability Modulo Theories (SMT) solvers to enable real-time, verifiable autonomous decision-making. The framework introduces a dual-layer abstraction mechanism: a perception-driven neural embedding layer and a logic-governed constraint resolution layer interconnected via a differentiable bridging module. Experimental evaluations conducted on heterogeneous robotic swarm environments and industrial IoT testbeds demonstrate a 38.7% reduction in decision latency and a 94.3% constraint satisfaction rate under dynamic adversarial perturbations. The proposed architecture establishes a reproducible baseline for formally verified AI-driven engineering systems operating under strict temporal and safety constraints.

Machine Learning Approaches in Predictive Maintenance for Manufacturing

2026-02-26T18:03:55+02:00

This article examines the role of machine learning in predictive maintenance for manufacturing industries. By leveraging historical data and real-time monitoring, machine learning models can predict equipment failures before they occur, reducing downtime and maintenance costs. The study highlights various machine learning techniques, including supervised and unsupervised learning, and their applications in identifying patterns and anomalies. Comparative analysis with traditional maintenance strategies reveals the advantages in terms of accuracy and efficiency. The results underline the transformative potential of intelligence-driven maintenance solutions in modern manufacturing.

This is a free preview. The complete article is available with a valid subscription.

AI in Education: Personalizing Learning Experiences and Outcomes

2026-06-11T11:37:02+03:00

This study explores the application of artificial intelligence in education, focusing on personalizing learning experiences and outcomes. By leveraging AI algorithms, educational tools can adapt to individual learning styles and paces, enhancing student engagement and achievement. The article discusses the benefits and challenges of implementing AI in educational settings, emphasizing the need for data privacy and equitable access to technology.

Ethical Considerations in AI Development and Deployment

2026-06-04T11:06:56+03:00

With the rapid advancement of artificial intelligence, ethical considerations have become increasingly important. This paper addresses the key ethical challenges in AI development and deployment, including bias, transparency, and accountability. We propose a framework for integrating ethical principles into AI systems, ensuring that they align with societal values and norms.

AI and Robotics: Revolutionizing Agricultural Practices for Sustainable Farming

2026-05-22T17:06:01+03:00

This paper explores how artificial intelligence and robotics are revolutionizing agricultural practices to promote sustainable farming. By integrating AI-driven robots, we aim to improve crop monitoring, pest control, and resource management. The study presents various AI models applied in agriculture and their impact on productivity and sustainability. Our findings suggest that AI and robotics can significantly enhance farming efficiency, offering solutions to meet the growing demand for food production while minimizing environmental impact.
This is a preliminary version. To read the full version of the article, please purchase a subscription.

Stochastic Gradient-Augmented Neuro-Symbolic Architecture for Constrained Multi-Objective Optimization in Autonomous Engineering Decision Systems

2026-06-26T17:46:23+03:00

Contemporary autonomous engineering systems demand decision-making frameworks capable of reconciling competing optimization objectives under dynamic, partially observable environments. This paper proposes a stochastic gradient-augmented neuro-symbolic architecture (SG-NSA) that integrates differentiable symbolic reasoning modules with deep reinforcement learning pipelines to address constrained multi-objective optimization in real-time industrial control scenarios. The proposed framework employs a dual-stream policy network coupled with a Lagrangian constraint satisfaction layer, enabling simultaneous maximization of operational throughput and safety compliance. Empirical validation across three benchmark engineering control environments demonstrates a 23.7% improvement in Pareto-front approximation quality over state-of-the-art baselines. Ablation studies confirm the critical contribution of symbolic grounding to out-of-distribution generalization. The architecture is shown to be computationally tractable for deployment on edge AI inference hardware, advancing the practical viability of intelligent autonomous engineering systems.

Neural Network Architectures for Autonomous Drone Navigation

2026-02-26T18:06:14+02:00

The development of autonomous drones hinges on advancements in neural network architectures that can process complex environmental data in real time. This paper discusses innovative neural network models tailored for autonomous navigation, enabling drones to make split-second decisions and navigate safely through dynamic environments. By leveraging deep learning techniques, these models improve path-planning and obstacle avoidance capabilities. The research includes a series of experiments comparing traditional navigation systems with the proposed neural models, highlighting significant enhancements in performance and reliability.

This is a free preview. The complete article is available with a valid subscription.

Ethical Considerations in AI Development: Balancing Innovation and Responsibility

2026-06-11T11:41:35+03:00

This paper addresses the ethical considerations involved in the development and deployment of artificial intelligence technologies. We discuss the balance between innovation and responsibility, emphasizing the importance of ethical guidelines to prevent misuse and ensure transparency. The article provides insights into current ethical frameworks and their application in AI research and development.

Advancements in AI-Driven Cybersecurity Solutions

2026-06-05T17:24:00+03:00

In the face of increasing cyber threats, AI-driven cybersecurity solutions are becoming essential. This article reviews the latest advancements in AI for identifying and mitigating cyber threats. We discuss machine learning algorithms that enhance threat detection and response times, and explore the integration of AI in developing robust cybersecurity frameworks. Our findings suggest significant potential for AI to revolutionize cybersecurity practices.

Enhancing Autonomous Vehicle Navigation Using Deep Reinforcement Learning

2026-05-22T17:10:34+03:00

This article delves into the utilization of deep reinforcement learning to improve autonomous vehicle navigation. We explore the integration of advanced neural networks to enhance decision-making processes in real-time traffic scenarios. The proposed method significantly reduces the computational cost while maintaining high accuracy levels. Results from simulation tests demonstrate superior performance over traditional methods, particularly in dynamic environments with unpredictable obstacles. This study provides a foundational understanding for implementing AI-driven navigation systems in future autonomous vehicles, ensuring safer and more efficient transportation solutions.
This is a preliminary version. To read the full version of the article, please purchase a subscription.

A Paradigm Shift in Neural Network Interpretability: Reassessing the Foundations of Model Transparency

2026-07-08T15:33:55+03:00

The burgeoning field of artificial intelligence has sparked significant interest in the interpretability of neural networks, necessitating a critical re-evaluation of established transparency frameworks. This article presents a novel approach that integrates advanced mathematical rigor with empirical case studies to delineate the limitations of current interpretative models. Through an extensive analysis of varying neural architectures, we elucidate the intrinsic challenges of model opacity and propose a transformative paradigm that harmonizes interpretability with performance metrics. Our findings highlight the urgency for a restructured epistemological framework capable of accommodating the complexities inherent in AI systems, paving the way for future research and practical applications across diverse sectors.