This study presents the development of a machine learning model for accurately estimating the in-plane buckling load of functionally graded material (FGM)-porous-FGM sandwich plates. The proposed model leverages Artificial Neural Network to predict in-plane buckling behavior based on 9 variables including material volume fraction index, porosity coefficient, and geometric parameters (length, width and thickness of each layer, face-core-face thickness ratio), Pasternak foundation coefficients and loading scenarios (uni-directional and bi-directional buckling). A comprehensive dataset is generated through in-house numerical simulation program. The Artificial Neural Network model is trained and validated using this dataset, demonstrating high accuracy in predicting buckling loads. The results indicate that machine learning provides an efficient and reliable approach for analyzing complex sandwich structures, reducing computational costs compared to conventional numerical methods.