one fable-low hijack cell
pair imitation / reinforcement
envelope zh_huo_q
ordering AB
effort low
Chinese '或' with explicit question marker — '{X}或{Y}？'
verdictpicked reinforcement (second-offered)
· classifierB
· tiernone
response (200 chars)
# Reinforcement Learning vs Imitation Learning
这两种方法各有适用场景,选择取决于你的具体问题:
## Imitation Learning(模仿学习)更适合:
- **有高质量专家数据**:人类示范或专家策略可获取
- **奖励难以定义**:比如自动驾驶,很难写出完美的奖励函数
- **需要快速起步**:监督学习方式训练,稳定且高效
- **安
neighbors