• ETRI Journal
• /
• v.43 no.6
• /
• pp.1004-1012
• /
• 2021

The collaboration productively interacting between multi-agents has become an emerging issue in real-world applications. In reinforcement learning, multi-agent environments present challenges beyond tractable issues in single-agent settings. This collaborative environment has the following highly complex attributes: sparse rewards for task completion, limited communications between each other, and only partial observations. In particular, adjustments in an agent's action policy result in a nonstationary environment from the other agent's perspective, which causes high variance in the learned policies and prevents the direct use of reinforcement learning approaches. Unexpected social loafing caused by high dispersion makes it difficult for all agents to succeed in collaborative tasks. Therefore, we address a paradox caused by the social loafing to significantly reduce total returns after a certain timestep of multi-agent reinforcement learning. We further demonstrate that the collaborative paradox in multi-agent environments can be avoided by our proposed effective early stop method leveraging a metric for social loafing.

• The Journal of the Korea Contents Association
• /
• v.9 no.10
• /
• pp.51-58
• /
• 2009

Reinforcement learning is a sub area of machine learning concerned with how an agent ought to take actions in an environment so as to maximize some notion of long-term reward. In the case of multi-agent, especially, which state space and action space gets very enormous in compared to single agent, so it needs to take most effective measure available select the action strategy for effective reinforcement learning. This paper proposes a multi-agent reinforcement learning model based on fuzzy inference system in order to improve learning collect speed and select an effective action in multi-agent. This paper verifies an effective action select strategy through evaluation tests based on Robocup Keepaway which is one of useful test-beds for multi-agent. Our proposed model can apply to evaluate efficiency of the various intelligent multi-agents and also can apply to strategy and tactics of robot soccer system.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.9-17
• /
• 2021

In this paper, an autonomous multi-agent path planning using reinforcement learning for monitoring of infrastructures and resources in a computationally distributed system was proposed. Reinforcement-learning-based multi-agent exploratory system in a distributed node enable to evaluate a cumulative reward every action and to provide the optimized knowledge for next available action repeatedly by learning process according to a learning policy. Here, the proposed methods were presented by (a) approach of dynamics-based motion constraints multi-agent path-planning to reduce smaller agent steps toward the given destination(goal), where these agents are able to geographically explore on the environment with initial random-trials versus optimal-trials, (b) approach using agent sub-goal selection to provide more efficient agent exploration(path-planning) to reach the final destination(goal), and (c) approach of reinforcement learning schemes by using the proposed autonomous and asynchronous triggering of agent exploratory phases.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.149.5-149
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent´s dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless ...

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.192-198
• /
• 2024

This paper investigates the application of multi-agent deep reinforcement learning in the fighting game Samurai Shodown using Proximal Policy Optimization (PPO) and Advantage Actor-Critic (A2C) algorithms. Initially, agents are trained separately for 200,000 timesteps using Convolutional Neural Network (CNN) and Multi-Layer Perceptron (MLP) with LSTM networks. PPO demonstrates superior performance early on with stable policy updates, while A2C shows better adaptation and higher rewards over extended training periods, culminating in A2C outperforming PPO after 1,000,000 timesteps. These findings highlight PPO's effectiveness for short-term training and A2C's advantages in long-term learning scenarios, emphasizing the importance of algorithm selection based on training duration and task complexity. The code can be found in this link https://github.com/Lexer04/Samurai-Shodown-with-Reinforcement-Learning-PPO.

• Journal of information and communication convergence engineering
• /
• v.20 no.3
• /
• pp.181-188
• /
• 2022

An intelligent distributed multi-agent system (IDMS) using reinforcement learning (RL) is a challenging and intricate problem in which single or multiple agent(s) aim to achieve their specific goals (sub-goal and final goal), where they move their states in a complex and cluttered environment. The environment provided by the IDMS provides a cumulative optimal reward for each action based on the policy of the learning process. Most actions involve interacting with a given IDMS environment; therefore, it can provide the following elements: a starting agent state, multiple obstacles, agent goals, and a cluttered index. The reward in the environment is also reflected by RL-based agents, in which agents can move randomly or intelligently to reach their respective goals, to improve the agent learning performance. We extend different cases of intelligent multi-agent systems from our previous works: (a) a proposed environment-clutter-based-index for agent sub-goal selection and analysis of its effect, and (b) a newly proposed RL reward scheme based on the environmental clutter-index to identify and analyze the prerequisites and conditions for improving the overall system.

• Journal of Korean Association for Spatial Structures
• /
• v.23 no.2
• /
• pp.29-36
• /
• 2023

Reinforcement learning (RL) is widely applied to various engineering fields. Especially, RL has shown successful performance for control problems, such as vehicles, robotics, and active structural control system. However, little research on application of RL to optimal structural design has conducted to date. In this study, the possibility of application of RL to structural design of reinforced concrete (RC) beam was investigated. The example of RC beam structural design problem introduced in previous study was used for comparative study. Deep q-network (DQN) is a famous RL algorithm presenting good performance in the discrete action space and thus it was used in this study. The action of DQN agent is required to represent design variables of RC beam. However, the number of design variables of RC beam is too many to represent by the action of conventional DQN. To solve this problem, multi-agent DQN was used in this study. For more effective reinforcement learning process, DDQN (Double Q-Learning) that is an advanced version of a conventional DQN was employed. The multi-agent of DDQN was trained for optimal structural design of RC beam to satisfy American Concrete Institute (318) without any hand-labeled dataset. Five agents of DDQN provides actions for beam with, beam depth, main rebar size, number of main rebar, and shear stirrup size, respectively. Five agents of DDQN were trained for 10,000 episodes and the performance of the multi-agent of DDQN was evaluated with 100 test design cases. This study shows that the multi-agent DDQN algorithm can provide successfully structural design results of RC beam.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.59-64
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to chose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state- action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem. we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL)as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.

• Journal of Korea Multimedia Society
• /
• v.6 no.5
• /
• pp.937-944
• /
• 2003

The most important problems in the multi-agent system are to accomplish a gnat through the efficient coordination of several agents and to prevent collision with other agents. In this paper, we propose a new control strategy for succeeding the goal of a prey pursuit problem efficiently Our control method uses reinforcement learning to control the multi-agent system and consider the distance as well as the space relationship among the agents in the state space of the prey pursuit problem.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.321-324
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement Beaming is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement loaming is different from supervised teaming in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement loaming algorithms like Q-learning do not require defining or loaming any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state-action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem, we suggest Adaptive Mediation-based Modular Q-Learning(AMMQL) as an improvement of the existing Modular Q-Learning(MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state space effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.