中文
相关论文

相关论文: An Analysis of Measure-Valued Derivatives for Poli…

200 篇论文

Reinforcement learning methods for robotics are increasingly successful due to the constant development of better policy gradient techniques. A precise (low variance) and accurate (low bias) gradient estimator is crucial to face…

机器学习 · 计算机科学 2021-07-21 João Carvalho , Davide Tateo , Fabio Muratore , Jan Peters

We present a new algorithm for stochastic variational inference that targets at models with non-differentiable densities. One of the key challenges in stochastic variational inference is to come up with a low-variance estimator of the…

机器学习 · 计算机科学 2018-10-26 Wonyeol Lee , Hangyeol Yu , Hongseok Yang

We present on-line policy gradient algorithms for computing the locally optimal policy of a constrained, average cost, finite state Markov Decision Process. The stochastic approximation algorithms require estimation of the gradient of the…

最优化与控制 · 数学 2018-12-18 Vikram Krishnamurthy , Felisa Vazquez Abad

We introduce a family of pairwise stochastic gradient estimators for gradients of expectations, which are related to the log-derivative trick, but involve pairwise interactions between samples. The simplest example of our new estimator,…

机器学习 · 计算机科学 2022-04-21 Christian J. Walder , Paul Roussel , Richard Nock , Cheng Soon Ong , Masashi Sugiyama

Variational inference using the reparameterization trick has enabled large-scale approximate Bayesian inference in complex probabilistic models, leveraging stochastic optimization to sidestep intractable expectations. The reparameterization…

机器学习 · 统计学 2020-02-13 Christian A. Naesseth , Francisco J. R. Ruiz , Scott W. Linderman , David M. Blei

Low-variance gradient estimation is crucial for learning directed graphical models parameterized by neural networks, where the reparameterization trick is widely used for those with continuous variables. While this technique gives…

机器学习 · 统计学 2016-11-07 Seiya Tokui , Issei sato

Policy gradient algorithms have proven to be successful in diverse decision making and control tasks. However, these methods suffer from high sample complexity and instability issues. In this paper, we address these challenges by providing…

机器学习 · 计算机科学 2021-03-17 Yannis Flet-Berliac , Reda Ouhamma , Odalric-Ambrym Maillard , Philippe Preux

The policy gradient approach is a flexible and powerful reinforcement learning method particularly for problems with continuous actions such as robot control. A common challenge in this scenario is how to reduce the variance of policy…

机器学习 · 计算机科学 2013-01-18 Tingting Zhao , Hirotaka Hachiya , Voot Tangkaratt , Jun Morimoto , Masashi Sugiyama

We derive an unbiased estimator for expectations over discrete random variables based on sampling without replacement, which reduces variance as it avoids duplicate samples. We show that our estimator can be derived as the…

机器学习 · 计算机科学 2020-02-17 Wouter Kool , Herke van Hoof , Max Welling

This paper considers policy search in continuous state-action reinforcement learning problems. Typically, one computes search directions using a classic expression for the policy gradient called the Policy Gradient Theorem, which decomposes…

机器学习 · 计算机科学 2020-04-13 Sujay Bhatt , Alec Koppel , Vikram Krishnamurthy

We propose policy gradient algorithms which learn risk-sensitive policies in a reinforcement learning (RL) framework. Our proposed algorithms maximize the distortion risk measure (DRM) of the cumulative reward in an episodic Markov decision…

机器学习 · 计算机科学 2024-02-06 Nithia Vijayan , Prashanth L. A

Gradient-based methods for optimisation of objectives in stochastic settings with unknown or intractable dynamics require estimators of derivatives. We derive an objective that, under automatic differentiation, produces low-variance…

机器学习 · 计算机科学 2019-09-25 Gregory Farquhar , Shimon Whiteson , Jakob Foerster

Policy gradient methods are reinforcement learning algorithms that adapt a parameterized policy by following a performance gradient estimate. Conventional policy gradient methods use Monte-Carlo techniques to estimate the gradient, which…

机器学习 · 计算机科学 2026-05-01 Mohammad Ghavamzadeh , Yaakov Engel , Michal Valko

Several authors have recently developed risk-sensitive policy gradient methods that augment the standard expected cost minimization problem with a measure of variability in cost. These studies have focused on specific risk-measures, such as…

人工智能 · 计算机科学 2015-06-09 Aviv Tamar , Yinlam Chow , Mohammad Ghavamzadeh , Shie Mannor

Reparameterization (RP) and likelihood ratio (LR) gradient estimators are used to estimate gradients of expectations throughout machine learning and reinforcement learning; however, they are usually explained as simple mathematical tricks,…

机器学习 · 计算机科学 2021-06-01 Paavo Parmas , Masashi Sugiyama

Off-policy stochastic actor-critic methods rely on approximating the stochastic policy gradient in order to derive an optimal policy. One may also derive the optimal policy by approximating the action-value gradient. The use of action-value…

机器学习 · 统计学 2017-03-14 Yemi Okesanjo , Victor Kofia

Policy gradient methods are very attractive in reinforcement learning due to their model-free nature and convergence guarantees. These methods, however, suffer from high variance in gradient estimation, resulting in poor sample efficiency.…

机器学习 · 计算机科学 2018-11-16 Sergey Pankov

Due to the high variance of policy gradients, on-policy optimization algorithms are plagued with low sample efficiency. In this work, we propose Augment-Reinforce-Merge (ARM) policy gradient estimator as an unbiased low-variance alternative…

机器学习 · 计算机科学 2019-03-14 Yunhao Tang , Mingzhang Yin , Mingyuan Zhou

Deterministic policy gradient algorithms are foundational for actor-critic methods in controlling continuous systems, yet they often encounter inaccuracies due to their dependence on the derivative of the critic's value estimates with…

机器学习 · 计算机科学 2025-02-11 Baturay Saglam , Dionysis Kalogerias

Optimization with noisy gradients has become ubiquitous in statistics and machine learning. Reparameterization gradients, or gradient estimates computed via the "reparameterization trick," represent a class of noisy gradients often used in…

机器学习 · 统计学 2017-05-23 Andrew C. Miller , Nicholas J. Foti , Alexander D'Amour , Ryan P. Adams
‹ 上一页 1 2 3 10 下一页 ›