# Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation

## TL;DR

This paper introduces the sliced Wasserstein discrepancy (SWD), a novel metric for unsupervised domain adaptation that aligns feature distributions by leveraging the Wasserstein metric and decision boundaries, improving performance across various vision tasks.

## Contribution

The paper proposes SWD, a new discrepancy measure combining Wasserstein metric and decision boundary information for effective unsupervised domain adaptation.

## Key findings

- Effective in digit and sign recognition tasks
- Improves image classification and semantic segmentation
- Applicable to object detection with consistent results

## Abstract

In this work, we connect two distinct concepts for unsupervised domain adaptation: feature distribution alignment between domains by utilizing the task-specific decision boundary and the Wasserstein metric. Our proposed sliced Wasserstein discrepancy (SWD) is designed to capture the natural notion of dissimilarity between the outputs of task-specific classifiers. It provides a geometrically meaningful guidance to detect target samples that are far from the support of the source and enables efficient distribution alignment in an end-to-end trainable fashion. In the experiments, we validate the effectiveness and genericness of our method on digit and sign recognition, image classification, semantic segmentation, and object detection.