Domain Adaptation and Entanglement: an Optimal Transport Perspective
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910869700476928 |
|---|---|
| author | Koç, Okan Soen, Alexander Chiang, Chao-Kai Sugiyama, Masashi |
| author_facet | Koç, Okan Soen, Alexander Chiang, Chao-Kai Sugiyama, Masashi |
| contents | Current machine learning systems are brittle in the face of distribution shifts (DS), where the target distribution that the system is tested on differs from the source distribution used to train the system. This problem of robustness to DS has been studied extensively in the field of domain adaptation. For deep neural networks, a popular framework for unsupervised domain adaptation (UDA) is domain matching, in which algorithms
try to align the marginal distributions in the feature or output space.
The current theoretical understanding of these methods, however, is limited and existing theoretical results are not precise enough to characterize their performance in practice.
In this paper, we derive new bounds based on optimal transport that analyze the UDA problem. Our new bounds include a term which we dub as \emph{entanglement}, consisting of an expectation of Wasserstein distance between conditionals with respect to changing data distributions. Analysis of the entanglement term provides a novel perspective on the unoptimizable aspects of UDA. In various experiments with multiple models across several DS scenarios, we show that this term can be used to explain the varying performance of UDA algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_08155 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Domain Adaptation and Entanglement: an Optimal Transport Perspective Koç, Okan Soen, Alexander Chiang, Chao-Kai Sugiyama, Masashi Machine Learning Current machine learning systems are brittle in the face of distribution shifts (DS), where the target distribution that the system is tested on differs from the source distribution used to train the system. This problem of robustness to DS has been studied extensively in the field of domain adaptation. For deep neural networks, a popular framework for unsupervised domain adaptation (UDA) is domain matching, in which algorithms try to align the marginal distributions in the feature or output space. The current theoretical understanding of these methods, however, is limited and existing theoretical results are not precise enough to characterize their performance in practice. In this paper, we derive new bounds based on optimal transport that analyze the UDA problem. Our new bounds include a term which we dub as \emph{entanglement}, consisting of an expectation of Wasserstein distance between conditionals with respect to changing data distributions. Analysis of the entanglement term provides a novel perspective on the unoptimizable aspects of UDA. In various experiments with multiple models across several DS scenarios, we show that this term can be used to explain the varying performance of UDA algorithms. |
| title | Domain Adaptation and Entanglement: an Optimal Transport Perspective |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2503.08155 |