# Tackling Partial Domain Adaptation with Self-Supervision

## TL;DR

This paper introduces a self-supervised approach using spatial co-location of patches to improve partial domain adaptation, effectively reducing negative transfer without relying on shared label spaces.

## Contribution

It reformulates a jigsaw puzzle self-supervision task for partial domain adaptation and demonstrates how it enhances existing adaptive methods.

## Key findings

- Improves domain adaptation performance on three datasets.
- Supports negative transfer reduction in partial domain settings.
- Enhances existing methods when combined with self-supervision.

## Abstract

Domain adaptation approaches have shown promising results in reducing the marginal distribution difference among visual domains. They allow to train reliable models that work over datasets of different nature (photos, paintings etc), but they still struggle when the domains do not share an identical label space. In the partial domain adaptation setting, where the target covers only a subset of the source classes, it is challenging to reduce the domain gap without incurring in negative transfer. Many solutions just keep the standard domain adaptation techniques by adding heuristic sample weighting strategies. In this work we show how the self-supervisory signal obtained from the spatial co-location of patches can be used to define a side task that supports adaptation regardless of the exact label sharing condition across domains. We build over a recent work that introduced a jigsaw puzzle task for domain generalization: we describe how to reformulate this approach for partial domain adaptation and we show how it boosts existing adaptive solutions when combined with them. The obtained experimental results on three datasets supports the effectiveness of our approach.