Signal Recovery from Random Dot-Product Graphs Under Local Differential Privacy
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916705542864896 |
|---|---|
| author | Vishwanath, Siddharth Hehir, Jonathan |
| author_facet | Vishwanath, Siddharth Hehir, Jonathan |
| contents | We consider the problem of recovering latent information from graphs under $\varepsilon$-edge local differential privacy where the presence of relationships/edges between two users/vertices remains confidential, even from the data curator. For the class of generalized random dot-product graphs, we show that a standard local differential privacy mechanism induces a specific geometric distortion in the latent positions. Leveraging this insight, we show that consistent recovery of the latent positions is achievable by appropriately adjusting the statistical inference procedure for the privatized graph. Furthermore, we prove that our procedure is nearly minimax-optimal under local edge differential privacy constraints. Lastly, we show that this framework allows for consistent recovery of geometric and topological information underlying the latent positions, as encoded in their persistence diagrams. Our results extend previous work from the private community detection literature to a substantially richer class of models and inferential tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_17274 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Signal Recovery from Random Dot-Product Graphs Under Local Differential Privacy Vishwanath, Siddharth Hehir, Jonathan Machine Learning Statistics Theory 68P27, 62H22, 62C20, 62R07 We consider the problem of recovering latent information from graphs under $\varepsilon$-edge local differential privacy where the presence of relationships/edges between two users/vertices remains confidential, even from the data curator. For the class of generalized random dot-product graphs, we show that a standard local differential privacy mechanism induces a specific geometric distortion in the latent positions. Leveraging this insight, we show that consistent recovery of the latent positions is achievable by appropriately adjusting the statistical inference procedure for the privatized graph. Furthermore, we prove that our procedure is nearly minimax-optimal under local edge differential privacy constraints. Lastly, we show that this framework allows for consistent recovery of geometric and topological information underlying the latent positions, as encoded in their persistence diagrams. Our results extend previous work from the private community detection literature to a substantially richer class of models and inferential tasks. |
| title | Signal Recovery from Random Dot-Product Graphs Under Local Differential Privacy |
| topic | Machine Learning Statistics Theory 68P27, 62H22, 62C20, 62R07 |
| url | https://arxiv.org/abs/2504.17274 |