TY  - RPRT
TI  - Principled Analysis of Deep Reinforcement Learning Evaluation and Design Paradigms
AU  - Ezgi Korkmaz
PY  - 2026
UR  - https://arxiv.org/abs/2607.07769
ID  - 2607.07769
ER  -