Search papers, labs, and topics across Lattice.
The paper introduces VAE-Inf, a two-stage framework for imbalanced classification that combines variational autoencoders (VAEs) for majority class representation with distribution-aware fine-tuning using limited minority samples. VAE-Inf constructs a Gaussian reference model of the majority class using Wasserstein barycenters of latent posteriors, then fine-tunes the encoder with a novel loss that enforces probabilistic separation based on variance-normalized projection statistics. The method achieves competitive performance and provides a statistically interpretable inference procedure with finite-sample Type-I error control.
Achieve robust imbalanced classification with scarce minority samples by turning a generative VAE into a discriminative classifier using distribution-aware fine-tuning and statistically sound hypothesis testing.
Imbalanced classification remains a pervasive challenge in machine learning, particularly when minority samples are too scarce to provide a robust discriminative boundary. In such extreme scenarios, conventional models often suffer from unstable decision boundaries and a lack of reliable error control. To bridge the gap between generative modeling and discriminative classification, we propose a two-stage framework \textbf{VAE-Inf} that integrates deep representation learning with statistically interpretable hypothesis testing. In the first stage, we adopt a one-class modeling perspective by training a variational autoencoder (VAE) exclusively on majority-class data to capture the underlying reference distribution. The resulting latent posteriors are aggregated via a Wasserstein barycenter to construct a global Gaussian reference model, providing a geometrically principled baseline for the majority class. In the second stage, we transform this generative foundation into a discriminative classifier by fine-tuning the encoder with limited minority samples. This is achieved through a novel distribution-aware loss that enforces probabilistic separation between classes based on variance-normalized projection statistics. For inference, we introduce a projection-based score that admits a natural hypothesis testing interpretation, allowing for a distribution-free calibration procedure. This approach yields exact finite-sample control of the Type-I error (false positive rate) without relying on restrictive parametric assumptions. Extensive experiments on diverse real-world benchmarks demonstrate that our framework achieves competitive performance against other approaches. The codes are available upon request.
