• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.2
• /
• pp.351-358
• /
• 2018

Reinforcement learning is a machine learning method which aims to determine a policy to get optimal actions in dynamic and stochastic environments. But reinforcement learning has high computational complexity and needs a lot of time to get solution, so it is not easily applicable to uncertain and continuous environments. To tackle the complexity problem, AC (actor-critic) method is used and it separates an action-value function into a value function and an action decision policy. Also, in transfer learning method, the knowledge constructed in one environment is adapted to another environment, so it reduces the time to learn in a reinforcement learning method. In this paper, we present AC method and transfer learning method to solve the problem of a reinforcement learning method. Finally, we analyze the case study which a transfer learning method is used to solve BS(base station) switching problem in wireless access networks.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.53 no.7
• /
• pp.517-524
• /
• 2004

For real world applications of reinforcement learning techniques, function approximation or generalization will be required to avoid curse of dimensionality. For this, an improved function approximation-based reinforcement teaming method is proposed to speed up convergence by using CA-CMAC(Credit-Assigned Cerebellar Model Articulation Controller). To show that our proposed CACRL(CA-CMAC-based Reinforcement Learning) performs better than the CRL(CMAC- based Reinforcement Learning), computer simulation and experiment results are illustrated, where a swing-up control Problem of an acrobot is considered.

• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.250-260
• /
• 2018

In order to control the quadcopter using reinforcement learning, hovering of 1-axial drones prototype is implemented through reinforcement learning. A complementary filter is used to measure the correct angle, and the range of angles is from -180 degrees to +180 degrees using modified complementary filter. The policy gradient method is used together with the REINFORCE algorithm for reinforcement learning. The prototype learned in this way confirmed the difference in performance depending on the length of the episode.

• 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.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.12 no.1
• /
• pp.910-917
• /
• 2020

In fire suppression, continuous delivery of water or foam to the fire source is essential. The present study concerns fire suppression in a ship under sea condition, by introducing reinforcement learning technique to aiming of fire extinguishing nozzle, which works in a ship compartment with six degrees of freedom movement by irregular waves. The physical modeling of the water jet and compartment motion was provided using Unity 3D engine. In the reinforcement learning, the change of the nozzle angle during the scenario was set as the action, while the reward is proportional to the ratio of the water particle delivered to the fire source area. The optimal control of nozzle aiming for continuous delivery of water jet could be derived. Various algorithms of reinforcement learning were tested to select the optimal one, the proximal policy optimization.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.217-221
• /
• 2023

최근 몇 년 동안, Large Language Model(LLM)의 발전은 인공 지능 연구 분야에서 주요 도약을 이끌어 왔다. 이러한 모델들은 복잡한 자연어처리 작업에서 뛰어난 성능을 보이고 있다. 특히 Human Alignment를 위해 Supervised Fine Tuning, Reinforcement Learning, Chain-of-Hindsight 등을 적용한 언어모델이 관심 받고 있다. 본 논문에서는 위에 언급한 3가지 지시학습 방법인 Supervised Fine Tuning, Reinforcement Learning, Chain-of-Hindsight 를 Llama, OPT 모델에 적용하여 성능을 측정 및 비교한다.

• Journal of IKEEE
• /
• v.25 no.4
• /
• pp.644-649
• /
• 2021

Self-Imitation Learning is a simple off-policy actor-critic algorithm that makes an agent find an optimal policy by using past good experiences. In case that Self-Imitation Learning is combined with reinforcement learning algorithms that have actor-critic architecture, it shows performance improvement in various game environments. However, its applications are limited to reinforcement learning algorithms that have actor-critic architecture. In this paper, we propose a method of applying Self-Imitation Learning to Deep Q-Network which is a value-based deep reinforcement learning algorithm and train it in various game environments. We also show that Self-Imitation Learning can be applied to Deep Q-Network to improve the performance of Deep Q-Network by comparing the proposed algorithm and ordinary Deep Q-Network training results.

• Journal of Internet Computing and Services
• /
• v.9 no.6
• /
• pp.165-177
• /
• 2008

An important issue in Multiagent reinforcement learning is how an agent should learn its optimal policy in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for Multiagent reinforcement learning tend to apply single-agent reinforcement learning techniques without any extensions or require some unrealistic assumptions even though they use explicit models of other agents. In this paper, a Naive Bayesian based policy model of the opponent agent is introduced and then the Multiagent reinforcement learning method using this model is explained. Unlike previous works, the proposed Multiagent reinforcement learning method utilizes the Naive Bayesian based policy model, not the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper, the Cat and Mouse game is introduced as an adversarial Multiagent environment. And then effectiveness of the proposed Naive Bayesian based policy model is analyzed through experiments using this game as test-bed.

• The Journal of Korea Robotics Society
• /
• v.15 no.4
• /
• pp.398-403
• /
• 2020

Robot's throwing control is difficult to accurately calculate because of air resistance and rotational inertia, etc. This complexity can be solved by using machine learning. Reinforcement learning using reward function puts limit on adapting to new environment for robots. Therefore, this paper applied deep reinforcement learning using neural network without reward function. Throwing is evaluated as a success or failure. AI network learns by taking the target position and control policy as input and yielding the evaluation as output. Then, the task is carried out by predicting the success probability according to the target location and control policy and searching the policy with the highest probability. Repeating this task can result in performance improvements as data accumulates. And this model can even predict tasks that were not previously attempted which means it is an universally applicable learning model for any new environment. According to the data results from 520 experiments, this learning model guarantees 75% success rate.

• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.721-736
• /
• 2021

For the mobile edge computing (MEC) system supporting dense network, a joint allocation algorithm of computing and communication resources based on reinforcement learning is proposed. The energy consumption of task execution is defined as the maximum energy consumption of each user's task execution in the system. Considering the constraints of task unloading, power allocation, transmission rate and calculation resource allocation, the problem of joint task unloading and resource allocation is modeled as a problem of maximum task execution energy consumption minimization. As a mixed integer nonlinear programming problem, it is difficult to be directly solve by traditional optimization methods. This paper uses reinforcement learning algorithm to solve this problem. Then, the Markov decision-making process and the theoretical basis of reinforcement learning are introduced to provide a theoretical basis for the algorithm simulation experiment. Based on the algorithm of reinforcement learning and joint allocation of communication resources, the joint optimization of data task unloading and power control strategy is carried out for each terminal device, and the local computing model and task unloading model are built. The simulation results show that the total task computation cost of the proposed algorithm is 5%-10% less than that of the two comparison algorithms under the same task input. At the same time, the total task computation cost of the proposed algorithm is more than 5% less than that of the two new comparison algorithms.