Probing the Frontiers of High-Energy Physics: Deep Learning-Based Invariant Mass Prediction from Electron-Electron Collisions

In this study, we analyzed eight deep learning models for predicting invariant mass in high-energy physics. By experimenting with various hyperparameters like learning rates, batch sizes, and hidden layers, we identified the best model with a remarkable Mean Squared Error (MSE) of 0.152689, showcasing superior performance within the range of 4 to 110. Our findings underscore the vital role of hyperparameter tuning in achieving optimal results, offering valuable insights for future research in high-energy physics model design and optimization.