Enforcing Self-Consistent Kinematic Constraints in Neutrino Energy Estimators

Machine learning algorithms have long been utilized across many experimental collaborations within the neutrino physics community in applications to ascertain singular kinematic quantities, such as initial neutrino energy for use in beam-based neutrino oscillation analyses, and perhaps angle in atmospheric experiments. However, most of these algorithms do not incorporate a coherent physical picture of initial neutrino kinematics, opting to introduce loss functions involving knowledge of only the relevant single variable, i.e. | p_ν |. Here, we argue for the introduction of composite (multiobjective) loss functions utilizing the full kinematic description of the neutrino, p_ν ≡ (E, p_x, p_y, p_z) compiling all relevant energy and angle information consistently. The use of such a fully defined variable can be seen as an example of Physics Informed Machine Learning. In this talk, we will review recent progress in understanding uses of these multiobjective loss functions in more consistently predicting kinematics for beam-like & atmospheric-like neutrinos, with associated metrics to evaluate improved performance over single variable examples.