Predicting Out-of-Distribution Error with Confidence Optimal Transport
This addresses the practical problem of when to trust deployed models facing distribution shifts, though it appears incremental as an application of optimal transport theory to a known bottleneck.
The paper tackles the problem of predicting machine learning model performance on out-of-distribution data without additional annotation, using Confidence Optimal Transport (COT) which achieves state-of-the-art results on three benchmark datasets and outperforms existing methods by a large margin.
Out-of-distribution (OOD) data poses serious challenges in deployed machine learning models as even subtle changes could incur significant performance drops. Being able to estimate a model's performance on test data is important in practice as it indicates when to trust to model's decisions. We present a simple yet effective method to predict a model's performance on an unknown distribution without any addition annotation. Our approach is rooted in the Optimal Transport theory, viewing test samples' output softmax scores from deep neural networks as empirical samples from an unknown distribution. We show that our method, Confidence Optimal Transport (COT), provides robust estimates of a model's performance on a target domain. Despite its simplicity, our method achieves state-of-the-art results on three benchmark datasets and outperforms existing methods by a large margin.