Parametric Information Maximization for Generalized Category Discovery
This work addresses category discovery in mixed labeled and unlabeled data, which is incremental but improves handling of class imbalance for practical applications.
The paper tackles the Generalized Category Discovery problem by introducing a Parametric Information Maximization model that mitigates class-balance bias, achieving new state-of-the-art performances across six datasets, particularly in fine-grained scenarios.
We introduce a Parametric Information Maximization (PIM) model for the Generalized Category Discovery (GCD) problem. Specifically, we propose a bi-level optimization formulation, which explores a parameterized family of objective functions, each evaluating a weighted mutual information between the features and the latent labels, subject to supervision constraints from the labeled samples. Our formulation mitigates the class-balance bias encoded in standard information maximization approaches, thereby handling effectively both short-tailed and long-tailed data sets. We report extensive experiments and comparisons demonstrating that our PIM model consistently sets new state-of-the-art performances in GCD across six different datasets, more so when dealing with challenging fine-grained problems.