Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

IPO: Interior-point Policy Optimization under Constraints

About

In this paper, we study reinforcement learning (RL) algorithms to solve real-world decision problems with the objective of maximizing the long-term reward as well as satisfying cumulative constraints. We propose a novel first-order policy optimization method, Interior-point Policy Optimization (IPO), which augments the objective with logarithmic barrier functions, inspired by the interior-point method. Our proposed method is easy to implement with performance guarantees and can handle general types of cumulative multiconstraint settings. We conduct extensive evaluations to compare our approach with state-of-the-art baselines. Our algorithm outperforms the baseline algorithms, in terms of reward maximization and constraint satisfaction.

Yongshuai Liu, Jiaxin Ding, Xin Liu• 2019

Related benchmarks

TaskDatasetResultRank
Constrained Reinforcement LearningGRID
Episodic Reward229.4
8
Constrained Reinforcement LearningHumanoid
Episodic Reward1.58e+3
8
Constrained Reinforcement LearningPointCircle
Episodic Reward68.7
8
Constrained Reinforcement LearningBottleneck
Episodic Reward279.3
8
Constrained Reinforcement LearningAntCircle
Episodic Reward149.3
8
Constrained Reinforcement LearningPointReach
Episodic Reward49.1
8
Constrained Reinforcement LearningAntReach
Episodic Reward45.2
8
Continuous ControlHalfCheetah-Velocity Safety-Gymnasium (test)
Reward1.82e+3
7
Safe Reinforcement LearningHopper-Velocity
Reward1.22e+3
7
Constrained Reinforcement LearningNavigation
Episodic Reward164.1
5
Showing 10 of 10 rows

Other info

Follow for update