Content-Style Learning from Unaligned Domains: Identifiability under Unknown Latent Dimensions

Understanding identifiability of latent content and style variables from unaligned multi-domain data is essential for tasks such as domain translation and data generation. Existing works on content-style identification were often developed under somewhat stringent conditions, e.g., that all latent components are mutually independent and that the dimensions of the content and style variables are known. We introduce a new analytical framework via cross-domain latent distribution matching (LDM), which establishes content-style identifiability under substantially more relaxed conditions. Specifically, we show that restrictive assumptions such as component-wise independence of the latent variables can be removed. Most notably, we prove that prior knowledge of the content and style dimensions is not necessary for ensuring identifiability, if sparsity constraints are properly imposed onto the learned latent representations. Bypassing the knowledge of the exact latent dimension has been a longstanding aspiration in unsupervised representation learning-our analysis is the first to underpin its theoretical and practical viability. On the implementation side, we recast the LDM formulation into a regularized multi-domain GAN loss with coupled latent variables. We show that the reformulation is equivalent to LDM under mild conditions-yet requiring considerably less computational resource. Experiments corroborate with our theoretical claims. INTRODUCTION In multi-domain learning, "domains" are typically characterized by a distinct "style" that sets their data apart from others (Choi et al., 2020) . Take handwritten digits as an example: writing styles of different persons can define different domains. Shared information across all domains, such as the identities of the digits in this case, is termed as "content". Learning content and style representations from multi-domain data facilitates many important applications, e.g., domain translation (