TY  - RPRT
TI  - A Geometric Perspective on Reward Function Updates in Inverse Reinforcement Learning
AU  - Anish Diwan
AU  - Jan Peters
AU  - Oleg Arenz
PY  - 2026
UR  - https://arxiv.org/abs/2610.14041
ID  - 2610.14041
ER  -