Go to ICLR 2025 Conference homepageOn Characterizing and Mitigating Imbalances in Multi-Instance Partial Label Learning
\textit{Multi-Instance Partial Label Learning} (MI-PLL) is a weakly-supervised learning setting encompassing \textit{partial label learning}, \textit{latent structural learning}, and \textit{neurosymbolic learning}. Unlike supervised learning, in MI-PLL, the inputs to the classifiers at training-time are tuples of instances $\mathbf{x}$. At the same time, the supervision signal is generated by a function $\sigma$ over the (hidden) gold labels of $\mathbf{x}$. In this work, we make multiple contributions towards addressing a problem that hasn’t been studied so far in the context of MI-PLL: that of characterizing and mitigating \textit{learning imbalances}, i.e., major differences in the errors occurring when classifying instances of different classes (aka \emph{class-specific risks}). In terms of theory, we derive class-specific risk bounds for MI-PLL, while making minimal assumptions. Our theory reveals a unique phenomenon: that $\sigma$ can greatly impact learning imbalances. This result is in sharp contrast with previous research on supervised and weakly-supervised learning, which only studies learning imbalances under the prism of data imbalances. On the practical side, we introduce a technique for estimating the marginal of the hidden labels using only MI-PLL data. Then, we introduce algorithms that mitigate imbalances at training- and testing-time, by treating the marginal of the hidden labels as a constraint. We demonstrate the effectiveness of our techniques using strong baselines from neurosymbolic and long-tail learning, suggesting performance improvements of up to 14%.