Differentially Private Learned Indexes
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909368307417088 |
|---|---|
| author | Du, Jianzhang Mudgal, Tilak Gadre, Rutvi Rahul Luo, Yukui Wang, Chenghong |
| author_facet | Du, Jianzhang Mudgal, Tilak Gadre, Rutvi Rahul Luo, Yukui Wang, Chenghong |
| contents | In this paper, we address the problem of efficiently answering predicate queries on encrypted databases, those secured by Trusted Execution Environments (TEEs), which enable untrusted providers to process encrypted user data without revealing its contents. A common strategy in modern databases to accelerate predicate queries is the use of indexes, which map attribute values (keys) to their corresponding positions in a sorted data array. This allows for fast lookup and retrieval of data subsets that satisfy specific predicates. Unfortunately, indexes cannot be directly applied to encrypted databases due to strong data dependent leakages. Recent approaches apply differential privacy (DP) to construct noisy indexes that enable faster access to encrypted data while maintaining provable privacy guarantees. However, these methods often suffer from large storage costs, with index sizes typically scaling linearly with the key space. To address this challenge, we propose leveraging learned indexes, a trending technique that repurposes machine learning models as indexing structures, to build more compact DP indexes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_21164 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Differentially Private Learned Indexes Du, Jianzhang Mudgal, Tilak Gadre, Rutvi Rahul Luo, Yukui Wang, Chenghong Databases Cryptography and Security Machine Learning In this paper, we address the problem of efficiently answering predicate queries on encrypted databases, those secured by Trusted Execution Environments (TEEs), which enable untrusted providers to process encrypted user data without revealing its contents. A common strategy in modern databases to accelerate predicate queries is the use of indexes, which map attribute values (keys) to their corresponding positions in a sorted data array. This allows for fast lookup and retrieval of data subsets that satisfy specific predicates. Unfortunately, indexes cannot be directly applied to encrypted databases due to strong data dependent leakages. Recent approaches apply differential privacy (DP) to construct noisy indexes that enable faster access to encrypted data while maintaining provable privacy guarantees. However, these methods often suffer from large storage costs, with index sizes typically scaling linearly with the key space. To address this challenge, we propose leveraging learned indexes, a trending technique that repurposes machine learning models as indexing structures, to build more compact DP indexes. |
| title | Differentially Private Learned Indexes |
| topic | Databases Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2410.21164 |