Solving Non-identifiable Latent Feature Models
This addresses a fundamental issue in statistical modeling for researchers and practitioners using latent feature models, though it appears incremental as it builds on existing methods as a post-process.
The paper tackles the problem of non-identifiability in latent feature models, which complicates parameter estimation due to equivalent solutions, by proposing a novel method that efficiently hops through these solutions to find an appropriate one, showing effectiveness on synthetic and real-world datasets.
Latent feature models (LFM)s are widely employed for extracting latent structures of data. While offering high, parameter estimation is difficult with LFMs because of the combinational nature of latent features, and non-identifiability is a particularly difficult problem when parameter estimation is not unique and there exists equivalent solutions. In this paper, a necessary and sufficient condition for non-identifiability is shown. The condition is significantly related to dependency of features, and this implies that non-identifiability may often occur in real-world applications. A novel method for parameter estimation that solves the non-identifiability problem is also proposed. This method can be combined as a post-process with existing methods and can find an appropriate solution by hopping efficiently through equivalent solutions. We have evaluated the effectiveness of the method on both synthetic and real-world datasets.