Gestalt Principles and Blast Domination Patterns in Neural Networks: A Conceptual Exploration
A. Ahila *
Department of Mathematics, Kalasalingam Academy of Research and Education – Deemed to be University, Virudhunagar, Tamil Nadu, India.
*Author to whom correspondence should be addressed.
Abstract
This research expounds an advanced integration of the Blast Domination Number, a graph-theoretic parameter quantifying control over network nodes, with neural network architectures to enhance learning efficiency and healthiness. By modelling neural networks as dynamic graphs, we identify optimal blast-dominating sets that influence neuron activations with minimal energy and maximum coverage. This novel approach enables targeted activation strategies, leading to improved convergence rates, reduced computational overhead, and superior performance in classification and pattern recognition tasks. Experimental results across standard datasets validate the method’s effectiveness, establishing a new paradigm in the application of combinatorial domination theory to adaptive neural computation and intelligent systems.
Keywords: Blast domination number, ANN (Artificial Neural Network), Python, neurons