The fatigue life of concrete under cyclic compression is a critical factor in assessing the durability and safety of concrete structures. In this study, we utilize machine learning techniques to predict the fatigue life, based on key material and stress parameters. The dataset, comprising samples of concrete tested under cyclic loading conditions, includes variables such as compressive strength (Fc), minimum and maximum cyclic stresses (Smin, Smax). We apply several regression models, including Linear Regression, Random Forest Regressor, and Gradient Boosting Regressor, to predict the fatigue life. Each model's performance is evaluated using R-squared (R2) metric. Preliminary findings indicate that ensemble methods, particularly Random Forest and Gradient Boosting, show superior predictive performance compared to traditional linear regression. These results underscore the potential of machine learning models to enhance fatigue life prediction.