Privacy-Preserving Data Deduplication for Enhancing Federated Learning of Language Models (Extended Version)
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929615603236864 |
|---|---|
| author | Abadi, Aydin Dasu, Vishnu Asutosh Sarkar, Sumanta |
| author_facet | Abadi, Aydin Dasu, Vishnu Asutosh Sarkar, Sumanta |
| contents | Deduplication is a vital preprocessing step that enhances machine learning model performance and saves training time and energy. However, enhancing federated learning through deduplication poses challenges, especially regarding scalability and potential privacy violations if deduplication involves sharing all clients' data. In this paper, we address the problem of deduplication in a federated setup by introducing a pioneering protocol, Efficient Privacy-Preserving Multi-Party Deduplication (EP-MPD). It efficiently removes duplicates from multiple clients' datasets without compromising data privacy. EP-MPD is constructed in a modular fashion, utilizing two novel variants of the Private Set Intersection protocol. Our extensive experiments demonstrate the significant benefits of deduplication in federated learning of large language models. For instance, we observe up to 19.62\% improvement in perplexity and up to 27.95\% reduction in running time while varying the duplication level between 10\% and 30\%. EP-MPD effectively balances privacy and performance in federated learning, making it a valuable solution for large-scale applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_08152 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Privacy-Preserving Data Deduplication for Enhancing Federated Learning of Language Models (Extended Version) Abadi, Aydin Dasu, Vishnu Asutosh Sarkar, Sumanta Cryptography and Security Artificial Intelligence Computation and Language Machine Learning Deduplication is a vital preprocessing step that enhances machine learning model performance and saves training time and energy. However, enhancing federated learning through deduplication poses challenges, especially regarding scalability and potential privacy violations if deduplication involves sharing all clients' data. In this paper, we address the problem of deduplication in a federated setup by introducing a pioneering protocol, Efficient Privacy-Preserving Multi-Party Deduplication (EP-MPD). It efficiently removes duplicates from multiple clients' datasets without compromising data privacy. EP-MPD is constructed in a modular fashion, utilizing two novel variants of the Private Set Intersection protocol. Our extensive experiments demonstrate the significant benefits of deduplication in federated learning of large language models. For instance, we observe up to 19.62\% improvement in perplexity and up to 27.95\% reduction in running time while varying the duplication level between 10\% and 30\%. EP-MPD effectively balances privacy and performance in federated learning, making it a valuable solution for large-scale applications. |
| title | Privacy-Preserving Data Deduplication for Enhancing Federated Learning of Language Models (Extended Version) |
| topic | Cryptography and Security Artificial Intelligence Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2407.08152 |