Model Rectification via Unknown Unknowns Extraction from Deployment Samples

TL;DR

This paper introduces RTSCV, a post-training framework that identifies and rectifies unknown unknowns in classifiers during deployment, significantly improving model performance without active learning.

Contribution

The paper proposes RTSCV, a novel algorithmic framework for model rectification that extracts unknown unknowns during deployment, with theoretical guarantees and superior empirical results.

Findings

RTSCV reduces performance gaps by up to 41%.

RTSCV outperforms state-of-the-art methods across datasets.

Theoretical guarantees support RTSCV's effectiveness.

Abstract

Model deficiency that results from incomplete training data is a form of structural blindness that leads to costly errors, oftentimes with high confidence. During the training of classification tasks, underrepresented class-conditional distributions that a given hypothesis space can recognize results in a mismatch between the model and the target space. To mitigate the consequences of this discrepancy, we propose Random Test Sampling and Cross-Validation (RTSCV) as a general algorithmic framework that aims to perform a post-training model rectification at deployment time in a supervised way. RTSCV extracts unknown unknowns (u.u.s), i.e., examples from the class-conditional distributions that a classifier is oblivious to, and works in combination with a diverse family of modern prediction models. RTSCV augments the training set with a sample of the test set (or deployment data) and uses this redefined class layout to discover u.u.s via cross-validation, without relying on active learning or budgeted queries to an oracle. We contribute a theoretical analysis that establishes performance guarantees based on the design bases of modern classifiers. Our experimental evaluation demonstrates RTSCV's effectiveness, using 7 benchmark tabular and computer vision datasets, by reducing a performance gap as large as 41% from the respective pre-rectification models. Last we show that RTSCV consistently outperforms state-of-the-art approaches.