Research Article
Theoretical Foundations of Neural Network Integration of System Software Modules
Evgeniy Bryndin*
Issue:
Volume 11, Issue 2, June 2025
Pages:
30-39
Received:
1 September 2025
Accepted:
12 September 2025
Published:
9 October 2025
Abstract: The theoretical foundations of neural network integration of system software modules lie in the field of combining formal methods for automation, optimization and management of development, integration and maintenance of complex software systems and systems engineering. The article proposes a formalism of operator schemes for constructing programs with deterministically connected modules for any class of algorithms. The structures of description and execution of programs are considered. In programs, there are data (information) and control transfer links between operators. When training neural networks, program structures indicate only the links that control their execution. Distributed links between inputs and outputs of sets of operators are associated with modules. Program operator modules are numbered. This numbering is preserved in program execution. Programs of modules can be sequential, parallel and sequential-parallel. The structure of programs with deterministically connected modules is considered. In the proposed formalism of operator schemes, the solvability of the problem of constructing programs with deterministically connected modules is proved. These fundamentals provide a theoretical basis for developing systems in which neural networks act as tools for complex automation and optimization of design, integration, and management of system software modules. In the future, the development of this concept will contribute to the creation of self-regulating, adaptive, and scalable system architectures.
Abstract: The theoretical foundations of neural network integration of system software modules lie in the field of combining formal methods for automation, optimization and management of development, integration and maintenance of complex software systems and systems engineering. The article proposes a formalism of operator schemes for constructing programs ...
Show More
Research Article
Implementation of AGI on Brain-Like Neuro-Network Structure
Evgeny Bryndin*
Issue:
Volume 11, Issue 2, June 2025
Pages:
40-48
Received:
19 September 2025
Accepted:
28 September 2025
Published:
12 November 2025
DOI:
10.11648/j.se.20251102.12
Downloads:
Views:
Abstract: The proposed approach to creating AGI using brain-like neural networks combines principles that teach systems to effectively generalize, remember, plan, and reason across a variety of tasks, drawing on ideas from neuromorphic architectures, dynamic hierarchical information processing, and hybrid neural-symbolic methods. Brain-like architectures dynamically process event-driven information over time, provide online learning and low power consumption, and implement research projects on hardware-based implementation of online plasticity. Within the framework of brain-like neural networks, this involves hybrid approaches to world reconstruction and action planning. The goal is to enable the network to reason, manipulate symbols, and use rules, similar to natural intelligence. This is achieved using differentiable logic, modal induction, neural memory processors, neural program interpreters, and neural-modular networks for task composition. This improves long-term memory, planning, and reasoning accuracy with a limited training set. It also enables skill transfer from one task to another, simulating alternative actions without multiple interactions with the real world. Brain-like neural networks have a modular, scalable architecture with routing and an ensemble of specialized modules. Neural modules for vision, planning, dialogue, and motor control can be dynamically assembled without complete retraining. Memory, planning, and perception are separated into separate modules with mechanisms for collaboration and joint goal learning. AGI based on brain-like neural networks enables knowledge transfer across domains. Neuromorphic modules motivate themselves to explore and investigate the environment, which supports long-term adaptability, similar to biological mechanisms. This is essential for AGI to function in the real world.
Abstract: The proposed approach to creating AGI using brain-like neural networks combines principles that teach systems to effectively generalize, remember, plan, and reason across a variety of tasks, drawing on ideas from neuromorphic architectures, dynamic hierarchical information processing, and hybrid neural-symbolic methods. Brain-like architectures dyn...
Show More
