Symmetry is a fundamental notion in mathematics and science, expressed through group actions that govern how objects transform consistently under symmetry operations. This concept, known as equivariance, plays a central role in various mathematical areas such as topology, geometry, representation theory, and K-theory by providing a framework to study structures while respecting their inherent symmetries. In machine learning, particularly in convolutional neural networks, equivariance formalizes how transformations of inputs lead to predictable transformations of outputs. Incorporating symmetry directly into network architectures enhances generalization, reduces the need for large training datasets, and improves robustness to transformations like rotations or reflections. Equivariant neural networks thus extend classical models by embedding symmetry as a design principle, offering superior performance in domains where data possess natural geometric or structural invariances, including computer vision, molecular modeling, and graph-based learning.