英语对照表

强化学习：reinforcement learning

智能体： agent

动作：action

决策：decision

奖励：reward

监督学习：supervised learning

损失函数：loss function

反向传播：back propagation

独立同分布：independent and identically distributed

延迟奖励：delayed reward

试错探索：trial-and-error exploration

探索：exploration

利用：exploitation

上限：upper bound

预演：rollout

轨迹：trajectory

回合：episode

序列决策：sequential decision making

完全可观测：fully observed

马尔可夫决策过程：markov decision process, MDP

部分可观测：partially observed

部分可观测的马尔可夫决策过程：partially observed markov decision process, POMDP

随机性策略：stochastic policy

确定性策略：deterministic policy