Abstract :
We present DeeLeMa, a deep learning network to analyze energies and momenta in particle collisions at high energy colliders, especially for collision events with multiple invisible particles. DeeLeMa is constructed based on symmetric event topology, and the generated mass distributions show robust peaks at the physical masses after the combinatoric uncertainties, and detector smearing effects are taken into account. DeeLeMa can be widely used in different event topologies by adopting the corresponding kinematic symmetries.