The paper proposes Maturing Markov Decision Processes (MMDPs) built on information-action asymmetry, an expiring-action priority principle, and a structure-aware RL framework, with experiments showing efficiency gains on replenishment, cash management, and production tasks.
Deep Reinforcement Learning in Parameterized Action Space
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
Recent work has shown that deep neural networks are capable of approximating both value functions and policies in reinforcement learning domains featuring continuous state and action spaces. However, to the best of our knowledge no previous work has succeeded at using deep neural networks in structured (parameterized) continuous action spaces. To fill this gap, this paper focuses on learning within the domain of simulated RoboCup soccer, which features a small set of discrete action types, each of which is parameterized with continuous variables. The best learned agent can score goals more reliably than the 2012 RoboCup champion agent. As such, this paper represents a successful extension of deep reinforcement learning to the class of parameterized action space MDPs.
years
2026 2representative citing papers
P-DQN with a hybrid discrete-continuous action space integrates CAV car-following and lane-changing and beats MOBIL+IDM on comfort and inverse-TTC in four simulated scenarios.
citing papers explorer
-
Maturing Markov Decision Processes: Decision Making under Increasing Information and Shrinking Action Sets
The paper proposes Maturing Markov Decision Processes (MMDPs) built on information-action asymmetry, an expiring-action priority principle, and a structure-aware RL framework, with experiments showing efficiency gains on replenishment, cash management, and production tasks.
-
Integrated Automated Car Following and Lane-changing control based on a Parametrized Deep Q-network with Hybrid Action Space
P-DQN with a hybrid discrete-continuous action space integrates CAV car-following and lane-changing and beats MOBIL+IDM on comfort and inverse-TTC in four simulated scenarios.