paper-with-me

Papers

Privately Aligning Language Models with Reinforcement Learning

2023-10-25 · Fan Wu, Huseyin A. Inan, Arturs Backurs, Varun Chandrasekaran, Janardhan Kulkarni, Robert Sim

Positioned between pre-training and user deployment, aligning large language models (LLMs) through reinforcement learning (RL) has emerged as a prevailing strategy for training instruction following-models such as ChatGPT. In this work, we initiate the study of privacy-preserving alignment of LLMs through Differential Privacy (DP) in conjunction with RL. Following the influential work of Ziegler et al. (2020), we study two dominant paradigms: (i) alignment via RL without human in the loop (e.g., positive review generation) and (ii) alignment via RL from human feedback (RLHF) (e.g., summarization in a human-preferred way). We give a new DP framework to achieve alignment via RL, and prove its correctness. Our experimental results validate the effectiveness of our approach, offering competitive utility while ensuring strong privacy protections.

📄 PDF Abstract BibTeX arXiv:2310.16960

Code (0)

등록된 구현이 없습니다.

Tasks

Instruction FollowingPrivacy Preservingreinforcement-learningReinforcement LearningReinforcement Learning (RL)Review Generation

Similar Papers 제목 키워드 기반

Privacy-Preserving Instructions for Aligning Large Language Models

2024-02-21 · Da Yu, Peter Kairouz, Sewoong Oh, Zheng Xu

Service providers of large language model (LLM) applications collect user instructions in the wild and use them in further aligning LLMs with users' intentions. These instructions, which potentially contain sensitive inf…

Language ModelingLanguage ModellingLarge Language ModelPrivacy Preserving

Privately generating tabular data using language models

2023-06-07 · Alexandre Sablayrolles, Yue Wang, Brian Karrer

Privately generating synthetic data from a table is an important brick of a privacy-first world. We propose and investigate a simple approach of treating each row in a table as a sentence and training a language model wi…

Language ModelingLanguage ModellingSentence

A Contextual-Bandit Oversight Game with Two-Sided Informational Asymmetry

2026-06-30 · Yunjin Tong arxiv

We study runtime human oversight of an AI agent when private information runs in both directions: the human privately knows her reward function, while the AI privately knows the quality of the action it proposes. This is…

Reinforcement Learning

Privately Learning Mixtures of Axis-Aligned Gaussians

2021-06-03 · NeurIPS 2021 12 · Ishaq Aden-Ali, Hassan Ashtiani, Christopher Liaw

We consider the problem of learning mixtures of Gaussians under the constraint of approximate differential privacy. We prove that $\widetilde{O}(k^2 d \log^{3/2}(1/\delta) / \alpha^2 \varepsilon)$ samples are sufficient …

Privately Customizing Prefinetuning to Better Match User Data in Federated Learning

2023-02-17 · Charlie Hou, Hongyuan Zhan, Akshat Shrivastava, Sid Wang 외

In Federated Learning (FL), accessing private client data incurs communication and privacy costs. As a result, FL deployments commonly prefinetune pretrained foundation models on a (large, possibly public) dataset that i…

Federated LearningLanguage ModelingLanguage ModellingLarge Language Model+1