paper-with-me

Papers

Discretizing Continuous Action Space for On-Policy Optimization

2019-01-29 · Yunhao Tang, Shipra Agrawal

In this work, we show that discretizing action space for continuous control is a simple yet powerful technique for on-policy optimization. The explosion in the number of discrete actions can be efficiently addressed by a policy with factorized distribution across action dimensions. We show that the discrete policy achieves significant performance gains with state-of-the-art on-policy optimization algorithms (PPO, TRPO, ACKTR) especially on high-dimensional tasks with complex dynamics. Additionally, we show that an ordinal parameterization of the discrete distribution can introduce the inductive bias that encodes the natural ordering between discrete actions. This ordinal architecture further significantly improves the performance of PPO/TRPO.

📄 PDF Abstract BibTeX arXiv:1901.10500

Code (2)

robintyh1/onpolicybaselines 공식 구현 tf
opendilab/DI-engine pytorch

Tasks

continuous-controlContinuous ControlInductive Bias

Methods 이 논문이 사용한 방법론

TRPO Trust Region Policy Optimization, or TRPO, is a policy gradient method in reinforcement learning that avoids parameter updates that change the policy too much with a KL…

Similar Papers 제목 키워드 기반

Discretizing Continuous Action Space with Unimodal Probability Distributions for On-Policy Reinforcement Learning

2024-08-01 · Yuanyang Zhu, Zhi Wang, Yuanheng Zhu, Chunlin Chen 외

For on-policy reinforcement learning, discretizing action space for continuous control can easily express multiple modes and is straightforward to optimize. However, without considering the inherent ordering between the …

continuous-controlContinuous Controlreinforcement-learningReinforcement Learning

Deep Jump Q-Evaluation for Offline Policy Evaluation in Continuous Action Space

2020-09-28 · Hengrui Cai, Chengchun Shi, Rui Song, Wenbin Lu

We consider off-policy evaluation (OPE) in continuous action domains, such as dynamic pricing and personalized dose finding. In OPE, one aims to learn the value under a new policy using historical data generated by a dif…

Off-policy evaluationQ-Learning

Discrete Sequential Prediction of Continuous Actions for Deep RL

2017-05-14 · ICLR 2018 1 · Luke Metz, Julian Ibarz, Navdeep Jaitly, James Davidson

It has long been assumed that high dimensional continuous control problems cannot be solved effectively by discretizing individual dimensions of the action space due to the exponentially large number of bins over which p…

continuous-controlContinuous ControlPredictionQ-Learning+1

Adaptive Discretization for Continuous Control using Particle Filtering Policy Network

2020-09-28 · Pei Xu, Ioannis Karamouzas

Controlling the movements of highly articulated agents and robots has been a long-standing challenge to model-free deep reinforcement learning. In this paper, we propose a simple, yet general, framework for improving the…

continuous-controlContinuous ControlDeep Reinforcement Learning

Bayesian Policy Optimization for Model Uncertainty

2018-10-01 · ICLR 2019 5 · Gilwoo Lee, Brian Hou, Aditya Mandalika, Jeongseok Lee 외

Addressing uncertainty is critical for autonomous systems to robustly adapt to the real world. We formulate the problem of model uncertainty as a continuous Bayes-Adaptive Markov Decision Process (BAMDP), where an agent …

model