paper-with-me

Papers

Deep Bayesian Quadrature Policy Optimization

2020-06-28 · Akella Ravi Tej, Kamyar Azizzadenesheli, Mohammad Ghavamzadeh, Anima Anandkumar, Yisong Yue

We study the problem of obtaining accurate policy gradient estimates using a finite number of samples. Monte-Carlo methods have been the default choice for policy gradient estimation, despite suffering from high variance in the gradient estimates. On the other hand, more sample efficient alternatives like Bayesian quadrature methods have received little attention due to their high computational complexity. In this work, we propose deep Bayesian quadrature policy gradient (DBQPG), a computationally efficient high-dimensional generalization of Bayesian quadrature, for policy gradient estimation. We show that DBQPG can substitute Monte-Carlo estimation in policy gradient methods, and demonstrate its effectiveness on a set of continuous control benchmarks. In comparison to Monte-Carlo estimation, DBQPG provides (i) more accurate gradient estimates with a significantly lower variance, (ii) a consistent improvement in the sample complexity and average return for several deep policy gradient algorithms, and, (iii) the uncertainty in gradient estimation that can be incorporated to further improve the performance.

📄 PDF Abstract BibTeX arXiv:2006.15637

Code (1)

Akella17/Deep-Bayesian-Quadrature-Policy-Optimization 공식 구현 pytorch

Tasks

continuous-controlContinuous ControlPolicy Gradient Methods

Similar Papers 제목 키워드 기반

SOBER: Highly Parallel Bayesian Optimization and Bayesian Quadrature over Discrete and Mixed Spaces

2023-01-27 · Masaki Adachi, Satoshi Hayakawa, Saad Hamid, Martin Jørgensen 외

Batch Bayesian optimisation and Bayesian quadrature have been shown to be sample-efficient methods of performing optimisation and quadrature where expensive-to-evaluate objective functions can be queried in parallel. How…

Bayesian OptimisationBayesian OptimizationDrug Discovery

Alternating Optimisation and Quadrature for Robust Control

2016-05-24 · Supratik Paul, Konstantinos Chatzilygeroudis, Kamil Ciosek, Jean-Baptiste Mouret 외

Bayesian optimisation has been successfully applied to a variety of reinforcement learning problems. However, the traditional approach for learning optimal policies in simulators does not utilise the opportunity to impro…

Bayesian OptimisationReinforcement Learning

Impact of Computation in Integral Reinforcement Learning for Continuous-Time Control

2024-02-27 · Wenhan Cao, Wei Pan

Integral reinforcement learning (IntRL) demands the precise computation of the utility function's integral at its policy evaluation (PEV) stage. This is achieved through quadrature rules, which are weighted sums of utili…

Exploiting gradients and Hessians in Bayesian optimization and Bayesian quadrature

2017-03-31 · Anqi Wu, Mikio C. Aoi, Jonathan W. Pillow

An exciting branch of machine learning research focuses on methods for learning, optimizing, and integrating unknown functions that are difficult or costly to evaluate. A popular Bayesian approach to this problem uses a …

Bayesian OptimizationGaussian Processes

Fast post-process Bayesian inference with Variational Sparse Bayesian Quadrature

2023-03-09 · Chengkun Li, Grégoire Clarté, Martin Jørgensen, Luigi Acerbi

In applied Bayesian inference scenarios, users may have access to a large number of pre-existing model evaluations, for example from maximum-a-posteriori (MAP) optimization runs. However, traditional approximate inferenc…

Active LearningBayesian InferenceVariational Inference