# SPOCC: Scalable POssibilistic Classifier Combination -- toward robust aggregation of classifiers

## TL;DR

This paper introduces SPOCC, a scalable possibilistic classifier aggregation method that combines predictions using possibility theory and adaptive t-norms, enhancing robustness in ensemble learning scenarios.

## Contribution

It presents a novel aggregation approach based on possibility theory and adaptive t-norms, improving robustness over traditional ensemble methods.

## Key findings

- The method effectively handles classifier dependencies and inaccuracies.
- It demonstrates robustness properties in classifier aggregation.
- The approach is scalable to large ensembles.

## Abstract

We investigate a problem in which each member of a group of learners is trained separately to solve the same classification task. Each learner has access to a training dataset (possibly with overlap across learners) but each trained classifier can be evaluated on a validation dataset. We propose a new approach to aggregate the learner predictions in the possibility theory framework. For each classifier prediction, we build a possibility distribution assessing how likely the classifier prediction is correct using frequentist probabilities estimated on the validation set. The possibility distributions are aggregated using an adaptive t-norm that can accommodate dependency and poor accuracy of the classifier predictions. We prove that the proposed approach possesses a number of desirable classifier combination robustness properties.