English
Related papers

Related papers: f-Divergence constrained policy improvement

200 papers

We derive a new variational formula for the R\'enyi family of divergences, $R_\alpha(Q\|P)$, between probability measures $Q$ and $P$. Our result generalizes the classical Donsker-Varadhan variational formula for the Kullback-Leibler…

Machine Learning · Statistics 2021-07-21 Jeremiah Birrell , Paul Dupuis , Markos A. Katsoulakis , Luc Rey-Bellet , Jie Wang

We prove that the $f$-divergences between univariate Cauchy distributions are all symmetric, and can be expressed as strictly increasing scalar functions of the symmetric chi-squared divergence. We report the corresponding scalar functions…

Information Theory · Computer Science 2022-12-26 Frank Nielsen , Kazuki Okamura

Reinforcement Learning from Human Feedback (RLHF) has become a cornerstone technique for post-training large language models. While most existing approaches rely on the reverse KL-regularization, recent empirical studies have begun…

Machine Learning · Computer Science 2026-05-11 Di Wu , Chengshuai Shi , Jing Yang , Cong Shen

Generative models frequently suffer miscalibration, wherein statistics of the sampling distribution, such as the fraction of generations in a given class, deviate from desired values. We frame calibration as a constrained optimization…

Machine Learning · Statistics 2026-05-29 Henry D. Smith , Nathaniel L. Diamant , Brian L. Trippe

Recently, there has been remarkable progress in reinforcement learning (RL) with general function approximation. However, all these works only provide regret or sample complexity guarantees. It is still an open question if one can achieve…

Machine Learning · Computer Science 2023-05-16 Yue Wu , Jiafan He , Quanquan Gu

Recently, a method called the Mutual Information Neural Estimator (MINE) that uses neural networks has been proposed to estimate mutual information and more generally the Kullback-Leibler (KL) divergence between two distributions. The…

Machine Learning · Computer Science 2019-08-20 Kartik Ahuja

In this paper, we use differential privacy as a lens to examine online learning in both full and partial information settings. The differential privacy framework is, at heart, less about privacy and more about algorithmic stability, and…

Machine Learning · Computer Science 2019-10-30 Jacob Abernethy , Young Hun Jung , Chansoo Lee , Audra McMillan , Ambuj Tewari

An abstract convergence theorem for a class of generalized descent methods that explicitly models relative errors is proved. The convergence theorem generalizes and unifies several recent abstract convergence theorems. It is applicable to…

Optimization and Control · Mathematics 2017-11-22 Peter Ochs

Proximal constraints are fundamental to the stability of the Large Language Model reinforcement learning. While the canonical clipping mechanism in PPO serves as an efficient surrogate for trust regions, we identify a critical bottleneck:…

Machine Learning · Computer Science 2026-03-06 Yuan Li , Bo Wang , Yufei Gao , Yuqian Yao , Xinyuan Wang , Zhangyue Yin , Xipeng Qiu

In this paper, we introduce new classes of divergences by extending the definitions of the Bregman divergence and the skew Jensen divergence. These new divergence classes (g-Bregman divergence and skew g-Jensen divergence) satisfy some…

Statistics Theory · Mathematics 2018-09-21 Tomohiro Nishiyama

Learning the value function of a given policy (target policy) from the data samples obtained from a different policy (behavior policy) is an important problem in Reinforcement Learning (RL). This problem is studied under the setting of…

Machine Learning · Computer Science 2019-11-14 Raghuram Bharadwaj Diddigi , Chandramouli Kamanchi , Shalabh Bhatnagar

As machine learning (ML) based systems are adopted in domains such as law enforcement, criminal justice, finance, hiring and admissions, ensuring the fairness of ML aided decision-making is becoming increasingly important. In this paper, we…

Machine Learning · Computer Science 2023-06-30 Meiyu Zhong , Ravi Tandon

Policy gradient is a widely utilized and foundational algorithm in the field of reinforcement learning (RL). Renowned for its convergence guarantees and stability compared to other RL algorithms, its practical application is often hindered…

Machine Learning · Computer Science 2024-04-12 Yunxiang Li , Rui Yuan , Chen Fan , Mark Schmidt , Samuel Horváth , Robert M. Gower , Martin Takáč

We interpret likelihood-based test functions from a geometric perspective where the Kullback-Leibler (KL) divergence is adopted to quantify the distance from a distribution to another. Such a test function can be seen as a sub-Gaussian…

Information Theory · Computer Science 2021-01-05 Yan Wang

Estimating Kullback-Leibler divergence from identical and independently distributed samples is an important problem in various domains. One simple and effective estimator is based on the k nearest neighbor distances between these samples.…

Information Theory · Computer Science 2020-02-27 Puning Zhao , Lifeng Lai

A key problem in reinforcement learning for control with general function approximators (such as deep neural networks and other nonlinear functions) is that, for many algorithms employed in practice, updates to the policy or $Q$-function…

Machine Learning · Computer Science 2016-03-01 Joshua Achiam

Policy gradient methods for Large Language Models optimize a policy $\pi_\theta$ via a surrogate objective computed from samples of a rollout policy $\pi_{\text{roll}}$. However, modern LLM-RL pipelines suffer from unavoidable…

Machine Learning · Computer Science 2026-03-02 Yingru Li , Jiacai Liu , Jiawei Xu , Yuxuan Tong , Ziniu Li , Qian Liu , Baoxiang Wang

Temporal difference (TD) learning is a foundational algorithm in reinforcement learning (RL). For nearly forty years, TD learning has served as a workhorse for applied RL as well as a building block for more complex and specialized…

Machine Learning · Computer Science 2025-06-24 Hwanwoo Kim , Panos Toulis , Eric Laber

Linear queries can be submitted to a server containing private data. The server provides a response to the queries systematically corrupted using an additive noise to preserve the privacy of those whose data is stored on the server. The…

Cryptography and Security · Computer Science 2020-08-12 Farhad Farokhi , Peyman Mohajerin Esfahani

We introduce Multi-CALF, an algorithm that intelligently combines reinforcement learning policies based on their relative value improvements. Our approach integrates a standard RL policy with a theoretically-backed alternative policy,…

Machine Learning · Computer Science 2025-05-20 Georgiy Malaniya , Anton Bolychev , Grigory Yaremenko , Anastasia Krasnaya , Pavel Osinenko
‹ Prev 1 8 9 10 Next ›