Switch-Based Multi-Part Neural Network

TL;DR

The paper tackles the limitations of centralized deep learning in decentralized and privacy-sensitive settings by introducing a switch-based multipart neural network in which neurons are trained on disjoint data slices and selectively activated by an input-driven switch. The core approach enables localized, brain-inspired learning with per-neuron specialization, modularity, and interpretability, suitable for edge computing and federated deployment. Key contributions include a dynamic switching mechanism, neuron-level data segmentation, independent localized training, and granular explainability via per-neuron analyses such as activation heatmaps. The proposed workflow—from data partitioning to centralized evaluation and visualization—demonstrates faster training cycles, improved transparency, and scalable deployment potential in privacy-preserving, distributed AI applications.

Abstract

This paper introduces decentralized and modular neural network framework designed to enhance the scalability, interpretability, and performance of artificial intelligence (AI) systems. At the heart of this framework is a dynamic switch mechanism that governs the selective activation and training of individual neurons based on input characteristics, allowing neurons to specialize in distinct segments of the data domain. This approach enables neurons to learn from disjoint subsets of data, mimicking biological brain function by promoting task specialization and improving the interpretability of neural network behavior. Furthermore, the paper explores the application of federated learning and decentralized training for real-world AI deployments, particularly in edge computing and distributed environments. By simulating localized training on non-overlapping data subsets, we demonstrate how modular networks can be efficiently trained and evaluated. The proposed framework also addresses scalability, enabling AI systems to handle large datasets and distributed processing while preserving model transparency and interpretability. Finally, we discuss the potential of this approach in advancing the design of scalable, privacy-preserving, and efficient AI systems for diverse applications.

Figures (4)

• Figure 1: Switch-Based Network Architecture
• Figure 2: Brain-Inspired Neural Network Architecture
• Figure 3: Process Flow of the Training Mechanism
• Figure 4: Neuron Activation Heatmap