Often in High-Energy Physics, large amounts of data are available for collisions with low transverse momentum (pT), but not for those with high pT. As a result, it is difficult to analyze data from collisions with high pT because there is not enough data to train a machine learning model effectively. This project tests the ability of neural networks to extrapolate to high pT data, having been trained on low pT data.
More info at (link)[https://www.divitrawal.com/research/#soft-extrap]