REVISTA
AI

   
Redirigiendo al acceso original de articulo en 17 segundos...
Inicio  /  AI  /  Vol: 2 Par: 4 (2021)  /  Artículo
ARTÍCULO
TITULO

Refined Continuous Control of DDPG Actors via Parametrised Activation

Mohammed Hossny    
Julie Iskander    
Mohamed Attia    
Khaled Saleh and Ahmed Abobakr    

Resumen

Continuous action spaces impose a serious challenge for reinforcement learning agents. While several off-policy reinforcement learning algorithms provide a universal solution to continuous control problems, the real challenge lies in the fact that different actuators feature different response functions due to wear and tear (in mechanical systems) and fatigue (in biomechanical systems). In this paper, we propose enhancing the actor-critic reinforcement learning agents by parameterising the final layer in the actor network. This layer produces the actions to accommodate the behaviour discrepancy of different actuators under different load conditions during interaction with the environment. To achieve this, the actor is trained to learn the tuning parameter controlling the activation layer (e.g., Tanh and Sigmoid). The learned parameters are then used to create tailored activation functions for each actuator. We ran experiments on three OpenAI Gym environments, i.e., Pendulum-v0, LunarLanderContinuous-v2, and BipedalWalker-v2. Results showed an average of 23.15% and 33.80% increase in total episode reward of the LunarLanderContinuous-v2 and BipedalWalker-v2 environments, respectively. There was no apparent improvement in Pendulum-v0 environment but the proposed method produces a more stable actuation signal compared to the state-of-the-art method. The proposed method allows the reinforcement learning actor to produce more robust actions that accommodate the discrepancy in the actuators? response functions. This is particularly useful for real life scenarios where actuators exhibit different response functions depending on the load and the interaction with the environment. This also simplifies the transfer learning problem by fine-tuning the parameterised activation layers instead of retraining the entire policy every time an actuator is replaced. Finally, the proposed method would allow better accommodation to biological actuators (e.g., muscles) in biomechanical systems.

 Artículos similares

       
 
Jan Stodt, Daniel Schönle, Christoph Reich, Fatemeh Ghovanlooy Ghajar, Dominik Welte and Axel Sikora    
In recent years, both the Internet of Things (IoT) and blockchain technologies have been highly influential and revolutionary. IoT enables companies to embrace Industry 4.0, the Fourth Industrial Revolution, which benefits from communication and connecti... ver más
Revista: Algorithms

 
Vaidas Luko?evicius, Rolandas Makaras and Andrius Dargu?is    
Two trends could be observed in the evolution of road transport. First, with the traffic becoming increasingly intensive, the motor road infrastructure is developed; more advanced, greater quality, and more durable materials are used; and pavement laying... ver más
Revista: Applied Sciences

 
Wessel Pienaar     Pág. 102 - 122
The commercial transportation of crude oil and petroleum products by pipeline is receiving increased attention in South Africa. Transnet Pipeline Transport has recently obtained permission from the National Energy Regulator of South Africa (Nersa)... ver más