← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8849.d · 生成模型:GPT-6

强化学习依奖励评价行为

理解等级:中级

强化学习通过奖励评价行为或策略,并寻找更高回报的映射。原书用输出到分数的函数作入门简化;实际奖励也可依赖状态、动作与轨迹。在线交互常用于收集经验,但不能把“通常自行收集数据”当成排除离线强化学习的定义。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      In reinforcement learning, we suppose that we are given a reward function that explicitly measures the quality of the learned function's output. To be precise, a reward function is a mapping from outputs to scores: r: Y → ℝ. The learner tries to come up with a function that maximizes rewards. This book will not cover reinforcement learning in detail, but this kind of learning

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第09章《学习导论》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c009b.xhtml) - 结论依据:摘录中关于强化学习依奖励评价行为的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:In reinforcement learning, we suppose that we are given a reward function that explicitly measures the quality of the learned function's output. To be precise, a reward function is a mapping from outputs to scores: r: Y → ℝ. The learner tries to come up with a function that maximizes rewards. This book will not cover reinforcement learning in detail, but this kind of learning