TY  - RPRT
TI  - Unified dynamics of training, learning, inference, and reinforcement in neural networks
AU  - Mian Wang
PY  - 2026
UR  - https://arxiv.org/abs/2611.01345
ID  - 2611.01345
ER  -