eCP: Equivariant Conformal Prediction with pre-trained models

TL;DR

This paper introduces eCP, a conformal prediction method enhanced with geometric symmetry information from pre-trained models, improving uncertainty quantification especially in high-confidence scenarios.

Contribution

The paper proposes a novel equivariant conformal prediction framework that leverages group symmetries to tighten uncertainty bounds in predictive models.

Findings

Theoretically guarantees contracted non-conformity scores with symmetry incorporation.

Improved exponential-tail bounds lead to sharper prediction sets.

Experimental validation in pedestrian trajectory prediction supports theoretical claims.

Abstract

Conformal prediction, a post-hoc, distribution-free, finite-sample method of uncertainty quantification that offers formal coverage guarantees under the assumption of data exchangeability. Unfortunately, the resulting uncertainty regions can grow significantly in long horizon missions, rendering the statistical guarantees uninformative. To that end, we propose infusing CP with geometric information via group-averaging of the pretrained predictor to distribute the non-conformity mass across the orbits. Each sample now is treated as a representative of an orbit, thus uncertainty can be mitigated by other samples entangled to it via the orbit inducing elements of the symmetry group. Our approach provably yields contracted non-conformity scores in increasing convex order, implying improved exponential-tail bounds and sharper conformal prediction sets in expectation, especially at high confidence levels. We then propose an experimental design to test these theoretical claims in pedestrian trajectory prediction.