Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.13350 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914374275301376 |
|---|---|
| author | Saoud, Abdel Djalil Sad Mboula, Fred Maurice Ngolè Slimani, Hanane |
| author_facet | Saoud, Abdel Djalil Sad Mboula, Fred Maurice Ngolè Slimani, Hanane |
| contents | Distributional shifts between training and inference time data remain a central challenge in machine learning, often leading to poor performance. It motivated the study of principled approaches for domain alignment, such as optimal transport based unsupervised domain adaptation, that relies on approximating Monge map using transport plans, which is sensitive to the transport problem regularization strategy and hyperparameters, and might yield biased domains alignment. In this work, we propose to interpret smoothed transport plans as adjacency matrices of bipartite graphs connecting source to target domain and derive domain-invariant samples' representations through spectral embedding. We evaluate our approach on acoustic adaptation benchmarks for music genre recognition, music-speech discrimination, as well as electrical cable defect detection and classification tasks using time domain reflection in different diagnosis settings, achieving overall strong performances. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_13350 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Beyond Mapping : Domain-Invariant Representations via Spectral Embedding of Optimal Transport Plans Saoud, Abdel Djalil Sad Mboula, Fred Maurice Ngolè Slimani, Hanane Machine Learning Distributional shifts between training and inference time data remain a central challenge in machine learning, often leading to poor performance. It motivated the study of principled approaches for domain alignment, such as optimal transport based unsupervised domain adaptation, that relies on approximating Monge map using transport plans, which is sensitive to the transport problem regularization strategy and hyperparameters, and might yield biased domains alignment. In this work, we propose to interpret smoothed transport plans as adjacency matrices of bipartite graphs connecting source to target domain and derive domain-invariant samples' representations through spectral embedding. We evaluate our approach on acoustic adaptation benchmarks for music genre recognition, music-speech discrimination, as well as electrical cable defect detection and classification tasks using time domain reflection in different diagnosis settings, achieving overall strong performances. |
| title | Beyond Mapping : Domain-Invariant Representations via Spectral Embedding of Optimal Transport Plans |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2601.13350 |