The current chapter explores the mathematical principles that form the basis of convolutional operations and their utilization within neural networks. We commence our analysis by examining the fundamental convolution operation and its significance in capturing localized patterns within datasets. Next, we proceed to discuss Convolutional Neural Networks (CNNs), emphasizing their capabilities in processing image data and extracting features. The narrative additionally encompasses Graph Neural Networks (GNNs), demonstrating their capacity to effectively process non-grid data structures and perform relational reasoning. In conclusion, we address the topic of convolutions in higher dimensions, highlighting their importance in the analysis of temporal and volumetric data. In conclusion, we will now proceed to establish the foundation for an examination of Recurrent Neural Networks (RNNs) and the subsequent integration of neural and symbolic approaches in the field of Neuro-Symbolic computing.