TY  - RPRT
TI  - Who breaks early, looses: goal oriented training of deep neural networks based on port Hamiltonian dynamics
AU  - Julian Burghoff
AU  - Marc Heinrich Monells
AU  - Hanno Gottschalk
PY  - 2023
UR  - https://arxiv.org/abs/2304.07070
ID  - 2304.07070
ER  -