English
Related papers

Related papers: f-Divergence constrained policy improvement

200 papers

In statistical classification/multiple hypothesis testing and machine learning, a model distribution estimated from the training data is usually applied to replace the unknown true distribution in the Bayes decision rule, which introduces a…

Information Theory · Computer Science 2024-09-24 Zijian Yang , Vahe Eminyan , Ralf Schlüter , Hermann Ney

The K-means algorithm is one of the most widely studied clustering algorithms in machine learning. While extensive research has focused on its ability to achieve a globally optimal solution, there still lacks a rigorous analysis of its…

Machine Learning · Computer Science 2025-06-12 Mingyi Li , Michael R. Metel , Akiko Takeda

The Pinsker inequality lower bounds the Kullback--Leibler divergence $D_{\textrm{KL}}$ in terms of total variation and provides a canonical way to convert $D_{\textrm{KL}}$ control into $\lVert \cdot \rVert_1$-control. Motivated by…

Information Theory · Computer Science 2026-02-06 Guglielmo Beretta , Tommaso Cesari , Roberto Colomboni

Kullback-Leibler (KL) divergence is a fundamental concept in information theory that quantifies the discrepancy between two probability distributions. In the context of Variational Autoencoders (VAEs), it serves as a central regularization…

Machine Learning · Computer Science 2026-04-14 Andrés Muñoz , Rodrigo Ramele

Black box variational inference (BBVI) with reparameterization gradients triggered the exploration of divergence measures other than the Kullback-Leibler (KL) divergence, such as alpha divergences. In this paper, we view BBVI with…

Machine Learning · Statistics 2018-01-09 Robert Bamler , Cheng Zhang , Manfred Opper , Stephan Mandt

We consider the nonlinear Kalman filtering problem using Kullback-Leibler (KL) and $\alpha$-divergence measures as optimization criteria. Unlike linear Kalman filters, nonlinear Kalman filters do not have closed form Gaussian posteriors…

Optimization and Control · Mathematics 2017-11-22 San Gultekin , John Paisley

Estimating the Kullback--Leibler (KL) divergence between language models has many applications, e.g., reinforcement learning from human feedback (RLHF), interpretability, and knowledge distillation. However, computing the exact KL…

Computation and Language · Computer Science 2025-10-28 Afra Amini , Tim Vieira , Ryan Cotterell

Reverse Kullback-Leibler (KL) divergence-based regularization with respect to a fixed reference policy is widely used in modern reinforcement learning to preserve the desired traits of the reference policy and sometimes to promote…

Machine Learning · Computer Science 2026-02-05 Anupam Nayak , Tong Yang , Osman Yagan , Gauri Joshi , Yuejie Chi

The $\alpha$-divergences include the well-known Kullback-Leibler divergence, Hellinger distance and $\chi^2$-divergence. In this paper, we derive differential and integral relations between the $\alpha$-divergences that are generalizations…

Information Theory · Computer Science 2022-11-29 Tomohiro Nishiyama

Recent Reinforcement Learning (RL) algorithms making use of Kullback-Leibler (KL) regularization as a core component have shown outstanding performance. Yet, only little is understood theoretically about why KL regularization helps, so far.…

Machine Learning · Computer Science 2021-01-07 Nino Vieillard , Tadashi Kozuno , Bruno Scherrer , Olivier Pietquin , Rémi Munos , Matthieu Geist

The logistic loss (a.k.a. cross-entropy loss) is one of the most popular loss functions used for multiclass classification. It is also the loss function of choice for next-token prediction in language modeling. It is associated with the…

Machine Learning · Computer Science 2025-06-16 Vincent Roulet , Tianlin Liu , Nino Vieillard , Michael E. Sander , Mathieu Blondel

Data processing inequalities for $f$-divergences can be sharpened using constants called "contraction coefficients" to produce strong data processing inequalities. For any discrete source-channel pair, the contraction coefficients for…

Information Theory · Computer Science 2018-07-17 Anuran Makur , Lizhong Zheng

It is commonly believed that optimizing the reverse KL divergence results in "mode seeking", while optimizing forward KL results in "mass covering", with the latter being preferred if the goal is to sample from multiple diverse modes. We…

Machine Learning · Computer Science 2025-10-24 Anthony GX-Chen , Jatin Prakash , Jeff Guo , Rob Fergus , Rajesh Ranganath

Neural Network (Deep Learning) is a modern model in Artificial Intelligence and it has been exploited in Survival Analysis. Although several improvements have been shown by previous works, training an excellent deep learning model requires…

Machine Learning · Statistics 2023-04-14 Li Liu , Xiangeng Fang , Di Wang , Weijing Tang , Kevin He

Density ratio estimation (DRE) is a core technique in machine learning used to capture relationships between two probability distributions. $f$-divergence loss functions, which are derived from variational representations of $f$-divergence,…

Machine Learning · Computer Science 2025-03-18 Yoshiaki Kitazawa

We propose a method to fuse posterior distributions learned from heterogeneous datasets. Our algorithm relies on a mean field assumption for both the fused model and the individual dataset posteriors and proceeds using a simple…

Machine Learning · Computer Science 2020-07-14 Sebastian Claici , Mikhail Yurochkin , Soumya Ghosh , Justin Solomon

We propose an extension of quasi-Newton methods, and investigate the convergence and the robustness properties of the proposed update formulae for the approximate Hessian matrix. Fletcher has studied a variational problem which derives the…

Computation · Statistics 2010-10-15 Takafumi Kanamori , Atsumi Ohara

We study the Unadjusted Langevin Algorithm (ULA) for sampling from a probability distribution $\nu = e^{-f}$ on $\mathbb{R}^n$. We prove a convergence guarantee in Kullback-Leibler (KL) divergence assuming $\nu$ satisfies a log-Sobolev…

Data Structures and Algorithms · Computer Science 2022-03-04 Santosh S. Vempala , Andre Wibisono

Deep nonlinear models pose a challenge for fitting parameters due to lack of knowledge of the hidden layer and the potentially non-affine relation of the initial and observed layers. In the present work we investigate the use of information…

Optimization and Control · Mathematics 2016-12-20 Jacob S. Hunter , Nathan O. Hodas

In reinforcement learning, temporal difference (TD) is the most direct algorithm to learn the value function of a policy. For large or infinite state spaces, exact representations of the value function are usually not available, and it must…

Machine Learning · Computer Science 2018-05-03 Yann Ollivier
‹ Prev 1 3 4 5 6 7 10 Next ›