变分自编码器的潜在空间结构
Variational autoencoders (VAEs) are widely used to build low-dimensional representations of scientific data, with the expectation that regularization yields smooth coordinates for interpolation, generation and optimization.
平滑性解释的局限性
However, applications in microscopy reveal domains, narrow transitions and rare-state islands in VAE latent spaces that smoothness alone does not explain.
与 Flory-Huggins 理论的对应关系
We show that, for prototype data with squared-error reconstruction and an optimal decoder, the latent-space objective has a direct correspondence with the Flory-Huggins theory of a regular solution, and use it to analyze how the objective, posterior distributions and latent dimensionality organize the representation.
域结构与后验坍缩
Data families occupy domains separated by finite-width walls that broaden with increasing regularization until class separation is lost through posterior collapse.
类质心在谐振阱中的行为
Class centroids behave like particles in a harmonic trap: the prior confines them, and structural dissimilarity sets a soft repulsion that determines which classes become neighbors and how far apart they sit, with separations growing only logarithmically with dissimilarity.
稀有状态的多种表现形式
Rare states can remain unresolved, form small islands, accumulate within walls or develop intervening domains, depending on their frequency and similarity to common states.
后验协方差与轴对齐
The posterior covariance can align domains with the latent axes, but axis alignment does not imply separation of physical factors.
旋转、切割与维度不足的影响
Rotations can produce rings and encoder cuts that reflect data symmetry, while insufficient dimensionality can lead to suppression, mixing or folding of additional factors.
与其他模型的比较
Comparisons with flow models, deep kernel learning (DKL) and hybrid VAE-DKL models show how these structures change when reconstruction and prediction play different roles.
意义与展望
We discuss implications for interpreting data, discovering rare states, generating intermediate states and optimizing properties.