← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6704.a · 生成模型:GPT-6

全局共现训练不等于奇异值分解

理解等级:中级

GloVe利用全局共现统计拟合词与上下文向量,使内积加偏置近似对数共现,并按频率加权误差。官方实现通过梯度更新求解,不是一次SVD,也不保证全局最优或同数据一定更快更准;原书将其直接等同SVD需纠正。

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

      Stanford NLP researchers,41 led by Jeffrey Pennington, set about to understand the reason why Word2Vec worked so well and to find the cost function that was being optimized. They started by counting the word co-occurrences and recording them in a square matrix. They found they could compute the singular value decomposition (SVD)42 of this co-occurrence matrix, splitting it into the same two weight matrices Word2Vec

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第06章《用词嵌入推理》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-6.html) - 结论依据:摘录中关于全局共现训练不等于奇异值分解的定义、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:Stanford NLP researchers,41 led by Jeffrey Pennington, set about to understand the reason why Word2Vec worked so well and to find the cost function that was being optimized. They started by counting the word co-occurrences and recording them in a square matrix. They found they could compute the singular value decomposition (SVD)42 of this co-occurrence matrix, splitting it into the same two weight matrices Word2Vec