Robust Flow-based Conformal Inference (FCI) with Statistical Guarantee

TL;DR

This paper introduces a flow-based conformal inference method that provides reliable confidence sets and outlier detection for high-dimensional, potentially contaminated data, overcoming exchangeability limitations.

Contribution

The paper proposes a novel flow-based conformal inference approach that is robust to contaminated testing data and preserves class-conditional distributions for uncertainty quantification.

Findings

Produces effective predictive sets

Accurately detects outliers

Outperforms competing methods

Abstract

Conformal prediction aims to determine precise levels of confidence in predictions for new objects using past experience. However, the commonly used exchangeable assumptions between the training data and testing data limit its usage in dealing with contaminated testing sets. In this paper, we develop a novel flow-based conformal inference (FCI) method to build predictive sets and infer outliers for complex and high-dimensional data. We leverage ideas from adversarial flow to transfer the input data to a random vector with known distributions. Our roundtrip transformation can map the input data to a low-dimensional space, meanwhile reserving the conditional distribution of input data given each class label, which enables us to construct a non-conformity score for uncertainty quantification. Our approach is applicable and robust when the testing data is contaminated. We evaluate our method, robust flow-based conformal inference, on benchmark datasets. We find that it produces effective predictive sets and accurate outlier detection and is more powerful relative to competing approaches.