Search

This paper proposes a novel reinforcement learning-based power allocation scheme, a unique approach to support QoS in MIMO systems. In MIMO systems, efficient power allocation is critical to mitigate inter-device interference and improve energy efficiency. Given the dynamic nature of the wireless channel environment and the potential changes in QoS levels, a dynamic channel allocation scheme that can satify the QoS with minimum power consumption is essential. Our proposed scheme, which allocates power to each antenna based on reinforcement learning, stands out for its ability to reduce unnecessary search range by varying the reward value according to the channel environment and QoS satisfaction estimated by pilot signals. Simulations further validate the feasibility and superiority of our innovative scheme. [ABSTRACT FROM AUTHOR]

As the types of application services available for mobile smart devices increase, the variability and irregularity of their battery run time also increases. It is difficult to expect good prediction accuracy when general regression methods exploiting the regularity of past battery run time data are applied to predict the battery run time. Therefore, various prediction strategies are needed according to the dynamic execution situations of mobile smart devices. A novel decision-making process is also required to learn and select a strategy suitable for the dynamic situation. In this paper, we propose a reinforcement learning based prediction framework that maximizes the reward for the action of selecting a prediction strategy appropriate for the situation while valuing all strategies with their rewards and carrying out the prediction strategy with the highest value for the next prediction action. The numerical results obtained by experimenting the proposed framework confirmed that it generated good performance of estimating battery run time. [ABSTRACT FROM AUTHOR]

The efficiency of container loading in logistics is a critical factor directly affecting a company’s profitability. Since improper loading planning can cause space inefficiency, product damage, additional costs, and work requirements, the importance of skilled workers who establish and carry out appropriate loading plans is emphasized. This 3D Bin Packing Problem (3D BPP) is an important task in various industries such as logistics, warehouse management, and cargo transportation. In this study, the latest reinforcement learning methodology is applied and effective solutions to 3D BPP are presented. In particular, this study uses the Proximal Policy Optimization (PPO) algorithm of reinforcement learning for stable and effective learning, learns Multimodal models and Graph Attention Networks (GAT) models, and evaluates loading capacity in 3D environments. In this paper, we experiment with two models through 3D loading environment and data under the same conditions and find the best model to apply practically to the field environment by deriving the optimal model. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel design approach for a wideband non-uniformly spaced linear array (NUSLA) to suppress the effect of interference signals. Notably, a uniform linear array (ULA), which is easy to handle, is widely utilized to generate nulls over a wide band; however, its beamforming performance is limited owing to the ULA structure. Although a NUSLA with nonlinear spacing addresses nonlinear problems that are difficult to handle mathematically, it can exhibit performance improvements surpassing those of ULA structures. However, an additional constraint is required to generate nulls at a specific position, and this has not yet been studied for null generation using wideband NUSLA. In this paper, we propose a novel cost function for designing a wideband NUSLA, which generates a null at the desired position, and we utilize the modified reinforcement learning algorithm (MORELA), which is a heuristic optimization algorithm based on reinforcement learning (RL), to minimize the proposed cost function and analyze the optimized antenna array and weights. Further, we compare the performance of the proposed MORELA based on RL with that of existing heuristic optimization algorithms via computer simulations. [ABSTRACT FROM AUTHOR]

In this paper, we propose a distributed routing algorithm for mobile ad hoc networks (MANET) where mobile devices can be utilized as relays for communication between remote source-destination nodes. The objective of the proposed algorithm is to minimize the end-to-end communication delay caused by transmission failure with deep channel fading. In each hop, the node needs to select the next relaying node by considering a tradeoff relationship between the link stability and forward link distance. Based on such feature, we formulate the problem with partially observable Markov decision process (MDP) and apply deep reinforcement learning to derive effective routing strategy for the formulated MDP. Simulation results show that the proposed algorithm outperforms other baseline schemes in terms of the average end-to-end delay. [ABSTRACT FROM AUTHOR]

In this paper, we suggest a reinforcement learning algorithm in satellite-aided communication system. Satellite-aided communication are drawing attention because it can increase stability and robustness of wireless communication. To improve the efficiency of satellite-aided communication, beamhopping method is introduced. In beamhopping, satellite beams are divided and operated in time-based system. The actual efficiency of beamhopping can be depend on the hopping pattern. To determine beamhopping pattern, we introduced multiplay Thompson sampling which is a family of reinforcement learning. We mathematically modeled the features of satellite communication and then applied reinforcement learning. By means of simulations, the performances of proposed learning were analyzed in terms of capacity and regret. [ABSTRACT FROM AUTHOR]

The MCTS algorithm is frequently used in game AI. The commonly used tree policy is the UCT algorithm. In this paper, we propose the use of MLPT (Multi-Level Priority List) algorithm as a tree policy to ensure performance and optimize it for different game fields. MLPT selects the first available node from the top-level children of the node and adjusts its priority based on the outcome. As a result, when starting bet was 2% of the current amount, the average reward ratio was 101.237%, standard deviation was 0.0815, and winning rate was 51.739%. In games played until bankruptcy, the winning rate was 97.8% with 47.7 wins and 60.2 losses. MLPT provided the most stable profits compared to other tree policies and showed that it was a stable bet with a higher number of wins. The proposed MLPT algorithm outperforms existing methods by providing stable bets. [ABSTRACT FROM AUTHOR]

In this paper, we study a scheduling problem based on reinforcement learning for overlay device-to-device (D2D) communication networks. Even though various technologies for D2D communication networks using Q-learning, which is one of reinforcement learning models, have been studied, Q-learning causes a tremendous complexity as the number of states and actions increases. In order to solve this problem, D2D communication technologies based on Deep Q Network (DQN) have been studied. In this paper, we thus design a DQN model by considering the characteristics of wireless communication systems, and propose a distributed scheduling scheme based on the DQN model that can reduce feedback and signaling overhead. The proposed model trains all parameters in a centralized manner, and transfers the final trained parameters to all mobiles. All mobiles individually determine their actions by using the transferred parameters. We analyze the performance of the proposed scheme by computer simulation and compare it with optimal scheme, opportunistic selection scheme and full transmission scheme. [ABSTRACT FROM AUTHOR]

Cognitive Radio Ad-Hoc Networks (CRAHNs) enable to overcome the shortage of frequency resources due to the increase of radio services. In order to avoid interference with the primary user in CRANH, channel sensing to check the idle channel is required, and when the primary user appears, the time delay due to handover should be minimized through fast idle channel selection. In this paper, throughput was improved by reducing the number of channel sensing and preferentially sensing a channel with a high probability of being idle, using reinforcement learning. In addition, we proposed a multi-channel MAC (Medium Access Control) protocol that can minimize the possibility of collision with the primary user by sensing the channel at the time of data transmission without performing periodic sensing. The performance was compared and analyzed through computer simulation. [ABSTRACT FROM AUTHOR]

This paper proposes reinforcement learning to solve the node-disjoint path problem which establishes multipath for reliable data transmission in wireless ad-hoc networks. The node-disjoint path problem is a problem of determining a plurality of paths so that the intermediate nodes do not overlap between the source and the destination. In this paper, we propose an optimization method considering transmission distance in a large-scale wireless ad-hoc network using Q-learning in reinforcement learning, one of machine learning. Especially, in order to solve the node-disjoint path problem in a large-scale wireless ad-hoc network, a large amount of computation is required, but the proposed reinforcement learning efficiently obtains appropriate results by learning the path. The performance of the proposed reinforcement learning is evaluated from the viewpoint of transmission distance to establish two node-disjoint paths. From the evaluation results, it showed better performance in the transmission distance compared with the conventional simulated annealing. [ABSTRACT FROM AUTHOR]

In this paper, we propose a design method for a busbar-PCB type EMI filter using the deep deterministic policy gradient (DDPG) algorithm, which is a reinforcement learning method. Owing to the presence of several parasitic elements in the busbar-PCB type EMI filter, the accuracy in the differential mode and common mode insertion loss in the wideband equivalent circuits was first demonstrated replacing the EM-circuit co-simulation. The state and behavior were defined, and learning was performed to satisfy the necessary insertion loss using the DDPG method. Consequently, impedances satisfying the required insertion loss were derived from the open impedance at five positions on the given PCB, and it was confirmed that the CISPR 25 class 4 regulation condition was satisfied in all frequency bands compared to that in the model used in the existing analysis. [ABSTRACT FROM AUTHOR]

In this paper, an antenna selection scheme is proposed in massive multi-user multiple input multiple output (MU-MIMO) systems. The proposed antenna selection scheme can achieve almost the same performance as a conventional scheme while significantly reducing the overhead of feedback by using deep reinforcement learning (DRL). Each user compares the channel gains of massive antennas in base station (BS) to the L-largest channel gain, converts them to one-bit binary numbers, and feed them back to BS. Thus, the feedback overhead can be significantly reduced. In the proposed scheme, DRL is adopted to prevent the performance loss that might be caused by the reduced feedback information. We carried out extensive Monte-Carlo simulations to analyze the performance of the proposed scheme and it was shown that the proposed scheme can achieve almost the same average sum-rates as a conventional scheme that is almost optimal. [ABSTRACT FROM AUTHOR]

Multi-path system is a system in which utilizes various networks simultaneously. It is expected that multi-path system can enhance communication speed, reliability, security of network. In this paper, we focus on path selection in multi-path system. To select optimal path, we propose deep reinforcement learning algorithm which is rewarded by the round-trip-time (RTT) of each networks. Unlike multi-armed bandit model, deep Q learning is applied to consider rapidly changing situations. Due to the delay of RTT data, we also suggest compensation algorithm of the delayed reward. Moreover, we implement testbed learning server to evaluate the performance of proposed algorithm. The learning server contains distributed database and tens or flow module to efficiently operate deep learning algorithm. By means of simulation, we showed that the proposed algorithm has better performance than lowest RTT about 20%. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

This study aims to develop a chatbot specialized for Korean empathy conversation based on task-oriented dialogue, generative language model, and reinforcement learning. This study fine-tuned the language model using the wellness dataset and the self-generated dataset through the ChatGPT API. After that, we created a dataset for reinforcement learning using human feedback. The performance of the chatbot was improved by tuning parameter through reinforcement learning. Experiments have shown that tuning language model's parameter through reinforcement learning can be effectively used in empathy conversations. It has also been shown that the improvement of the performance of chatbots provides empathetic dialogue to users who need emotional support. [ABSTRACT FROM AUTHOR]