Post-Hoc Domain Adaptation via Guided Data Homogenization

TL;DR

This paper introduces a post-hoc domain adaptation method called guided data homogenization that adjusts data distributions to enable safe deployment of models in new environments without modifying the trained model.

Contribution

It proposes a novel data-centric approach to domain adaptation that leverages implicit training data information to transfer models to new domains without retraining.

Findings

Effective in shifting data distributions on CIFAR-10 and MNIST

Enables deployment in new scenarios without model retraining

Preserves safety certifications of deployed models

Abstract

Addressing shifts in data distributions is an important prerequisite for the deployment of deep learning models to real-world settings. A general approach to this problem involves the adjustment of models to a new domain through transfer learning. However, in many cases, this is not applicable in a post-hoc manner to deployed models and further parameter adjustments jeopardize safety certifications that were established beforehand. In such a context, we propose to deal with changes in the data distribution via guided data homogenization which shifts the burden of adaptation from the model to the data. This approach makes use of information about the training data contained implicitly in the deep learning model to learn a domain transfer function. This allows for a targeted deployment of models to unknown scenarios without changing the model itself. We demonstrate the potential of data homogenization through experiments on the CIFAR-10 and MNIST data sets.