On Margins and Generalisation for Voting Classifiers
This work provides theoretical insights into ensemble learning, addressing a foundational debate in machine learning, though it is incremental as it builds on prior methods like Dirichlet posteriors.
The paper tackles the problem of understanding generalization in majority voting classifiers by proving margin-based generalization bounds using PAC-Bayes theory, achieving state-of-the-art guarantees on classification tasks.
We study the generalisation properties of majority voting on finite ensembles of classifiers, proving margin-based generalisation bounds via the PAC-Bayes theory. These provide state-of-the-art guarantees on a number of classification tasks. Our central results leverage the Dirichlet posteriors studied recently by Zantedeschi et al. [2021] for training voting classifiers; in contrast to that work our bounds apply to non-randomised votes via the use of margins. Our contributions add perspective to the debate on the "margins theory" proposed by Schapire et al. [1998] for the generalisation of ensemble classifiers.