本頁只刊出中文翻譯與中文說明；英文原文請見下方原文連結。
原文連結
論文資訊
- 類型：已發表論文
- 日期：2023-10-16
摘要
Variational autoencoders and Helmholtz machines use a recognition 網絡 (encoder) to approximate the posterior distribution of a generative model (decoder). In this paper we establish some necessary and some sufficient properties of a recognition 網絡 so that it can model the true posterior distribution exactly. These results are derived in the general context of probabilistic graphical modelling / 貝氏 網絡s, for which the 網絡 represents a set of conditional independence statements. We derive both global conditions, in terms of d-separation, and local conditions for the recognition 網絡 to have the desired qualities. It turns out that for the local conditions the perfectness property (for every node, all parents are joined) plays an important role.
※ 此為已發表論文，全文需透過期刊付費取得