English
Related papers

Related papers: Differentially Private Reward Estimation with Pref…

200 papers

Standard Bradley--Terry (BT) reward models are limited when human preferences are pluralistic. Although soft preference labels preserve disagreement information, BT can only express it by shrinking reward margins. Gaussian reward models…

Machine Learning · Statistics 2026-05-13 Shuxing Fang , Ruijian Han , Liangyu Zhang , Fan Zhou

In the era of increasing privacy concerns and demand for personalized experiences, traditional Reinforcement Learning with Human Feedback (RLHF) frameworks face significant challenges due to their reliance on centralized data. We introduce…

Machine Learning · Computer Science 2025-02-11 Flint Xiaofeng Fan , Cheston Tan , Yew-Soon Ong , Roger Wattenhofer , Wei-Tsang Ooi

Federated learning (FL) allows to train a massive amount of data privately due to its decentralized structure. Stochastic gradient descent (SGD) is commonly used for FL due to its good empirical performance, but sensitive user information…

Machine Learning · Computer Science 2021-02-10 Muah Kim , Onur Günlü , Rafael F. Schaefer

Existing alignment methods share a common topology of information flow, where reward information is collected from humans, modeled with preference learning, and used to tune language models. However, this shared topology has not been…

Machine Learning · Computer Science 2025-05-29 Tianyi Qiu , Fanzhi Zeng , Jiaming Ji , Dong Yan , Kaile Wang , Jiayi Zhou , Yang Han , Josef Dai , Xuehai Pan , Yaodong Yang

Reinforcement Learning with Verifiable reward (RLVR) on preference data has become the mainstream approach for training Generative Reward Models (GRMs). Typically in pairwise rewarding tasks, GRMs generate reasoning chains ending with…

Computation and Language · Computer Science 2026-05-04 Zongqi Wang , Rui Wang , Yuchuan Wu , Yiyao Yu , Pinyi Zhang , Shaoning Sun , Yujiu Yang , Yongbin Li

Train machine learning models on sensitive user data has raised increasing privacy concerns in many areas. Federated learning is a popular approach for privacy protection that collects the local gradient information instead of real data.…

Cryptography and Security · Computer Science 2021-05-24 Lichao Sun , Jianwei Qian , Xun Chen

Reinforcement learning from human feedback (RLHF) implicitly aggregates heterogeneous human preferences into a single utility function, even though the underlying utilities of the participants are in practice diverse. Hence, RLHF can be…

Computer Science and Game Theory · Computer Science 2026-01-28 Zhiyu An , Duaa Nakshbandi , Wan Du

In recent years, Local Differential Privacy (LDP), a robust privacy-preserving methodology, has gained widespread adoption in real-world applications. With LDP, users can perturb their data on their devices before sending it out for…

Machine Learning · Computer Science 2023-08-02 Héber H. Arcolezi , Karima Makhlouf , Catuscia Palamidessi

With the rapid advances in Large Language Models (LLMs), aligning LLMs with human preferences become increasingly important. Although Reinforcement Learning with Human Feedback (RLHF) proves effective, it is complicated and highly…

Computation and Language · Computer Science 2024-10-31 Shiqi Wang , Zhengze Zhang , Rui Zhao , Fei Tan , Cam Tu Nguyen

Large language models are increasingly used for patient-facing medical assistance and clinical decision support, but adapting them to clinical dialogue often requires supervision derived from doctor-patient conversations that may contain…

Computation and Language · Computer Science 2026-03-10 Sudip Bhujel

Sensitive statistics are often collected across sets of users, with repeated collection of reports done over time. For example, trends in users' private preferences or software usage may be monitored via such reports. We study the…

Machine Learning · Computer Science 2020-07-28 Úlfar Erlingsson , Vitaly Feldman , Ilya Mironov , Ananth Raghunathan , Kunal Talwar , Abhradeep Thakurta

Current RLHF methods such as PPO and DPO typically reduce human preferences to binary labels, which are costly to obtain and too coarse to reflect individual variation. We observe that expressions of satisfaction and dissatisfaction follow…

Computation and Language · Computer Science 2025-10-28 YuXuan Zhang

Differential Privacy (DP) is an important privacy-enhancing technology for private machine learning systems. It allows to measure and bound the risk associated with an individual participation in a computation. However, it was recently…

Machine Learning · Computer Science 2022-09-09 Cuong Tran , My H. Dinh , Ferdinando Fioretto

Motivated by the increasing deployment of reinforcement learning in the real world, involving a large consumption of personal data, we introduce a differentially private (DP) policy gradient algorithm. We show that, in this setting, the…

Machine Learning · Computer Science 2025-02-03 Alexandre Rio , Merwan Barlier , Igor Colin

Reinforcement Learning from Human Feedback (RLHF) has played a crucial role in the success of large models such as ChatGPT. RLHF is a reinforcement learning framework which combines human feedback to improve learning effectiveness and…

Machine Learning · Computer Science 2023-11-28 Feiyang Han , Yimin Wei , Zhaofeng Liu , Yanxing Qi

We consider a refinement of differential privacy --- per instance differential privacy (pDP), which captures the privacy of a specific individual with respect to a fixed data set. We show that this is a strict generalization of the standard…

Machine Learning · Statistics 2018-11-15 Yu-Xiang Wang

Large language models (LLMs) trained on web-scale corpora can memorize sensitive training data, posing significant privacy risks. Differential privacy (DP) has emerged as a principled framework that limits the influence of individual data…

Computation and Language · Computer Science 2026-05-13 Eduardo Tenorio , Karuna Bhaila , Xintao Wu

Designing privacy-preserving machine learning algorithms has received great attention in recent years, especially in the setting when the data contains sensitive information. Differential privacy (DP) is a widely used mechanism for data…

Machine Learning · Computer Science 2025-09-11 Chunyang Liao , Deanna Needell , Hayden Schaeffer , Alexander Xue

This paper proposes new methodologies for conducting practical differentially private (DP) estimation and inference in high-dimensional linear regression. We first introduce a DP Bayesian Information Criterion (DP-BIC) for selecting the…

Methodology · Statistics 2026-04-13 Zhanrui Cai , Sai Li , Xintao Xia , Linjun Zhang

In the recent decades, the advance of information technology and abundant personal data facilitate the application of algorithmic personalized pricing. However, this leads to the growing concern of potential violation of privacy due to…

Machine Learning · Statistics 2021-09-13 Xi Chen , Sentao Miao , Yining Wang