Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Ma, Olivia, Passerat-Palmbach, Jonathan, Usynin, Dmitrii
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866915032472748032
author Ma, Olivia
Passerat-Palmbach, Jonathan
Usynin, Dmitrii
author_facet Ma, Olivia
Passerat-Palmbach, Jonathan
Usynin, Dmitrii
contents Fine-tuning large language models (LLMs) for specific tasks introduces privacy risks, as models may inadvertently memorise and leak sensitive training data. While Differential Privacy (DP) offers a solution to mitigate these risks, it introduces significant computational and performance trade-offs, particularly with standard fine-tuning approaches. Previous work has primarily focused on full-parameter updates, which are computationally intensive and may not fully leverage DPs potential in large models. In this work, we address these shortcomings by investigating Parameter-Efficient Fine-Tuning (PEFT) methods under DP constraints. We show that PEFT methods achieve comparable performance to standard fine-tuning while requiring fewer parameters and significantly reducing privacy leakage. Furthermore, we incorporate a data poisoning experiment involving intentional mislabelling to assess model memorisation and directly measure privacy risks. Our findings indicate that PEFT methods not only provide a promising alternative but also serve as a complementary approach for privacy-preserving, resource-efficient fine-tuning of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2411_15831
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models
Ma, Olivia
Passerat-Palmbach, Jonathan
Usynin, Dmitrii
Machine Learning
Artificial Intelligence
Fine-tuning large language models (LLMs) for specific tasks introduces privacy risks, as models may inadvertently memorise and leak sensitive training data. While Differential Privacy (DP) offers a solution to mitigate these risks, it introduces significant computational and performance trade-offs, particularly with standard fine-tuning approaches. Previous work has primarily focused on full-parameter updates, which are computationally intensive and may not fully leverage DPs potential in large models. In this work, we address these shortcomings by investigating Parameter-Efficient Fine-Tuning (PEFT) methods under DP constraints. We show that PEFT methods achieve comparable performance to standard fine-tuning while requiring fewer parameters and significantly reducing privacy leakage. Furthermore, we incorporate a data poisoning experiment involving intentional mislabelling to assess model memorisation and directly measure privacy risks. Our findings indicate that PEFT methods not only provide a promising alternative but also serve as a complementary approach for privacy-preserving, resource-efficient fine-tuning of LLMs.
title Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2411.15831