• Annual Conference of KIPS
• /
• 2004.11a
• /
• pp.525-528
• /
• 2004

먹이추적문제(prey pursuit problem)는 가상 격자로 이루어진 공간 내에 다중의 에이전트를 이용하여 먹이를 포획하는 것이다. 에이전트들은 먹이를 포획하기 위해 $30{\times}30$으로 이루어진 격자공간 (gride)안에서 기존 제안된 지역 제어, 분산 제어, 강화학습을 이용한 분산 제어 전략들을 적용하여 먹이를 포획하는 전략을 구현하였다. 제한된 격자 공간은 현실세계를 표현하기에는 너무도 역부족이어서 본 논문에서는 제한된 격자공간이 아닌 현실 세계와 흡사한 무한 공간 환경을 표현하고자 하였다. 표현된 환경의 모델은 순환구조(circular)형 격자 공간이라는 새로운 실험 공간이며, 새로운 공간에 맞는 전략은 에이전트와 먹이와의 추적 관계를 방향 벡터를 고려한 모델로 구현하였다. 기존 실험과는 차별화 된 환경에서 에이전트들은 휴리스틱을 통한 학습을 할 수 있다는 가정과 먹이의 효율적 포획, 충돌문제 해결이라는 결과를 얻었다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.205-211
• /
• 2024

The purpose of this study is to implement an agent that intelligently performs tracking and movement through reinforcement learning using the Unity and ML-Agents. In this study, we conducted an experiment to compare the learning performance between training one agent in a single learning simulation environment and parallel training of several agents simultaneously in a multi-learning simulation environment. From the experimental results, we could be confirmed that the parallel training method is about 4.9 times faster than the single training method in terms of learning speed, and more stable and effective learning occurs in terms of learning stability.

• Annual Conference of KIPS
• /
• 2019.05a
• /
• pp.493-495
• /
• 2019

제어 분야의 가장 기초적인 시스템인 Rotary Inverted Pendulum 을 제어하기 위하여, 본 논문에서는 강화학습에서 Deep Q-Network 과 함께 대표적인 알고리즘으로 알려진 Asynchronous Advantage Actor-Critic 을 활용하여 다중 디바이스 제어를 설계한다. Deep Q-Network 알고리즘을 활용한 기존 연구와 동일한 방식으로 실 세계의 물리 에이전트와 가상 환경을 맵핑시키며, 스위치를 통하여 로컬 에이전트와 글로벌 네트워크 간 통신을 구성한다. 본 논문에서는 분산 Asynchronous Advantage Actor-Critic 을 이용함으로써 실 세계의 다중 에이전트 제어를 위한 강화 학습의 활용 가능성을 조명한다.

• Journal of Digital Convergence
• /
• v.19 no.12
• /
• pp.347-352
• /
• 2021

In this paper, we propose a method for finding feature subsets that are effective for classification in an input dataset by using a multi-agent reinforcement learning method. In the field of machine learning, it is crucial to find features suitable for classification. A dataset may have numerous features; while some features may be effective for classification or prediction, others may have little or rather negative effects on results. In machine learning problems, feature selection for increasing classification or prediction accuracy is a critical problem. To solve this problem, we proposed a feature selection method based on reinforced learning. Each feature has one agent, which determines whether the feature is selected. After obtaining corresponding rewards for each feature that is selected, but not by the agents, the Q-value of each agent is updated by comparing the rewards. The reward comparison of the two subsets helps agents determine whether their actions were right. These processes are performed as many times as the number of episodes, and finally, features are selected. As a result of applying this method to the Wisconsin Breast Cancer, Spambase, Musk, and Colon Cancer datasets, accuracy improvements of 0.0385, 0.0904, 0.1252 and 0.2055 were shown, respectively, and finally, classification accuracies of 0.9789, 0.9311, 0.9691 and 0.9474 were achieved, respectively. It was proved that our proposed method could properly select features that were effective for classification and increase classification accuracy.

• Annual Conference of KIPS
• /
• 2020.11a
• /
• pp.1034-1035
• /
• 2020

인공지능은 스스로 학습하며 기존 통계 분석보다 탁월한 분석 역량을 지니고 있어 스마트팩토리 혁신에 새로운 전기를 마련할 것으로 기대된다. 이를 증명하듯 스마트팩토리의 주요 분야인 공정 간 연계 제어, 전문가 공정 제어, 로봇 자동화 등에서 활발한 연구가 이어지고 있다. 본 논문에서는 소포물 분류 시스템에 전통적인 룰 기반의 제어 방식 대신 다중 에이전트 강화 학습 제어 방식을 설계 및 적용하여 효과적인 행동 제어가 가능함을 입증한다.

• Journal of KIISE:Software and Applications
• /
• v.30 no.7_8
• /
• pp.672-680
• /
• 2003

A shopbot is a software agent whose goal is to maximize buyer´s satisfaction through automatically gathering the price and quality information of goods as well as the services from on-line sellers. In the response to shopbots´ activities, sellers on the Internet need the agents called pricebots that can help them maximize their own profits. In this paper we adopts Q-learning, one of the model-free reinforcement learning methods as a price-setting algorithm of pricebots. A Q-learned agent increases profitability and eliminates the cyclic price wars when compared with the agents using the myoptimal (myopically optimal) pricing strategy Q-teaming needs to select a sequence of state-action fairs for the convergence of Q-teaming. When the uniform random method in selecting state-action pairs is used, the number of accesses to the Q-tables to obtain the optimal Q-values is quite large. Therefore, it is not appropriate for universal on-line learning in a real world environment. This phenomenon occurs because the uniform random selection reflects the uncertainty of exploitation for the optimal policy. In this paper, we propose a Mixed Nonstationary Policy (MNP), which consists of both the auxiliary Markov process and the original Markov process. MNP tries to keep balance of exploration and exploitation in reinforcement learning. Our experiment results show that the Q-learning agent using MNP converges to the optimal Q-values about 2.6 time faster than the uniform random selection on the average.

• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.207-212
• /
• 2004

This paper presents a reinforcement learning framework for stock trading systems. Trading system parameters are optimized by Q-learning algorithm and neural networks are adopted for value approximation. In this framework, cooperative multiple agents are used to efficiently integrate global trend prediction and local trading strategy for obtaining better trading performance. Agents Communicate With Others Sharing training episodes and learned policies, while keeping the overall scheme of conventional Q-learning. Experimental results on KOSPI 200 show that a trading system based on the proposed framework outperforms the market average and makes appreciable profits. Furthermore, in view of risk management, the system is superior to a system trained by supervised learning.

• Journal of the Korea Society for Simulation
• /
• v.27 no.1
• /
• pp.1-13
• /
• 2018

Earthwork planning is one of the critical issues in a construction process management. For the construction process management, there are some different approaches such as optimizing construction with either mathematical methodologies or heuristics with simulations. This paper propose a simulated earthwork scenario and an optimal path for the simulation using a reinforcement learning. For reinforcement learning, we use two different Markov decision process, or MDP, formulations with interacting excavator agent and truck agent, sequenced learning, and independent learning. The simulation result shows that two different formulations can reach the optimal planning for a simulated earthwork scenario. This planning could be a basis for an automatic construction management.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.8
• /
• pp.171-180
• /
• 2020

Smart Factory consists of digital automation solutions throughout the production process, including design, development, manufacturing and distribution, and it is an intelligent factory that installs IoT in its internal facilities and machines to collect process data in real time and analyze them so that it can control itself. The smart factory's equipment works in a physical combination of numerous hardware, rather than a virtual character being driven by a single object, such as a game. In other words, for a specific common goal, multiple devices must perform individual actions simultaneously. By taking advantage of the smart factory, which can collect process data in real time, if reinforcement learning is used instead of general machine learning, behavior control can be performed without the required training data. However, in the real world, it is impossible to learn more than tens of millions of iterations due to physical wear and time. Thus, this paper uses simulators to develop grid sortation systems focusing on transport facilities, one of the complex environments in smart factory field, and design cooperative multi-agent-based reinforcement learning to demonstrate efficient behavior control.

• Annual Conference of KIPS
• /
• 2020.05a
• /
• pp.585-586
• /
• 2020

공정 데이터를 실시간으로 수집할 수 있는 스마트 팩토리의 장점을 활용하여, 일반적인 기계 학습 대신 강화 학습을 사용한다면 미리 요구되는 훈련 데이터 없이 행동 제어를 할 수 있다. 하지만, 현실 세계에서는 물리적 마모, 시간적 문제 등으로 인해 수천만 번 이상의 반복 학습이 불가능하다. 따라서, 본 논문에서는 시뮬레이터를 활용해 스마트 팩토리 분야에서 복잡한 환경 중 하나인 이송 설비에 초점을 둔 그리드 분류 시스템을 개발하고 협력적 다중 에이전트 기반의 강화 학습을 설계하여 효율적인 행동 제어가 가능함을 입증한다.