110,805 Hits in 2.1 sec

A View on Deep Reinforcement Learning in System Optimization [article]

Ameer Haj-Ali, Nesreen K. Ahmed, Ted Willke, Joseph Gonzalez, Krste Asanovic, Ion Stoica
2019 arXiv   pre-print
In this paper, we review, discuss, and evaluate the recent trends of using deep reinforcement learning in system optimization.  ...  We propose a set of essential metrics to guide future works in evaluating the efficacy of using deep reinforcement learning in system optimization.  ...  ., 2019) that focus on machine learning methods without discussing deep RL models or applying them beyond a specific system problem, we focus on deep RL in system optimization in general.  ... 
arXiv:1908.01275v3 fatcat:ih52psaazzcs3pulz4nnnjk2di

Deep Reinforcement Learning Based Volt-VAR Optimization in Smart Distribution Systems [article]

Ying Zhang, Xinan Wang, Jianhui Wang, Yingchen Zhang
2020 arXiv   pre-print
This paper develops a model-free volt-VAR optimization (VVO) algorithm via multi-agent deep reinforcement learning (MADRL) in unbalanced distribution systems.  ...  This method is novel since we cast the VVO problem in unbalanced distribution networks to an intelligent deep Q-network (DQN) framework, which avoids solving a specific optimization model directly when  ...  To address the limitations of these model-based approaches, recent effort applies reinforcement learning (RL) to power system operation, such as voltage control [17] - [19] .  ... 
arXiv:2003.03681v2 fatcat:kaysoexl3rg7fjmpwdjwb5ki3q

Deep Reinforcement Learning for Practical Phase Shift Optimization in RIS-aided MISO URLLC Systems [article]

Ramin Hashemi, Samad Ali, Nurul Huda Mahmood, Matti Latva-aho
2022 arXiv   pre-print
The considered scenario is a finite blocklength (FBL) regime and the problem is solved by leveraging a novel deep reinforcement learning (DRL) algorithm named twin-delayed deep deterministic policy gradient  ...  system.  ...  In recent years, machine learning methods, particularly deep reinforcement learning (DRL) algorithms, have been considered as a reliable and powerful framework in wireless communications [12] , [13]  ... 
arXiv:2110.08513v3 fatcat:2twsllrvybbapnd4nrygl6ksce

Deep reinforcement learning for optimal well control in subsurface systems with uncertain geology [article]

Yusuf Nasir, Louis J. Durlofsky
2022 arXiv   pre-print
A general control policy framework based on deep reinforcement learning (DRL) is introduced for closed-loop decision making in subsurface flow settings.  ...  The DRL-based methodology is shown to result in an NPV increase of 15% (for the 2D cases) and 33% (3D cases) relative to robust optimization over prior models, and to an average improvement of 4% in NPV  ...  We acknowledge the Stanford Center for Computational Earth & Environmental Science for providing the computational resources used in this study.  ... 
arXiv:2203.13375v1 fatcat:z6xarlspinedlokhzp65o4idqq

Deep Reinforcement Learning for Joint Beamwidth and Power Optimization in mmWave Systems [article]

Jiabao Gao, Caijun Zhong, Xiaoming Chen, Hai Lin, Zhaoyang Zhang
2020 arXiv   pre-print
This paper studies the joint beamwidth and transmit power optimization problem in millimeter wave communication systems. A deep reinforcement learning based approach is proposed.  ...  Specifically, a customized deep Q network is trained offline, which is able to make real-time decisions when deployed online.  ...  Therefore, we propose to use deep reinforcement learning (DRL), a mechanism which does not require labels naturally.  ... 
arXiv:2006.13518v1 fatcat:owidsqmp3jcxjminb6o2y7coz4

Deep Reinforcement Learning for Real-Time Optimization of Pumps in Water Distribution Systems

Gergely Hajgató, György Paál, Bálint Gyires-Tóth
2020 Journal of water resources planning and management  
Deep reinforcement learning (DRL) is presented here as a controller of pumps in two WDSs.  ...  Real-time control of pumps can be an infeasible task in water distribution systems (WDSs) because the calculation to find the optimal pump speeds is resource-intensive.  ...  The research presented in this paper has been supported by the BME-Artificial Intelligence  ... 
doi:10.1061/(asce)wr.1943-5452.0001287 fatcat:vy6wgrwhiragdgtshjragc3iii

Optimized Computation Offloading Performance in Virtual Edge Computing Systems via Deep Reinforcement Learning [article]

Xianfu Chen, Honggang Zhang, Celimuge Wu, Shiwen Mao, Yusheng Ji, Mehdi Bennis
2018 arXiv   pre-print
To break the curse of high dimensionality in state space, we first propose a double deep Q-network (DQN) based strategic computation offloading algorithm to learn the optimal policy without knowing a priori  ...  Nevertheless, the design of computation offloading policies for a virtual MEC system remains challenging.  ...  APPROACHING THE OPTIMAL POLICY In this section, we proceed to approach the optimal control policy by developing practically feasible algorithms based on recent advances in deep reinforcement learning and  ... 
arXiv:1805.06146v1 fatcat:w7bty6jqdfgdborxp2gtorq5uy

Power Allocation in Cache-Aided NOMA Systems: Optimization and Deep Reinforcement Learning Approaches [article]

Khai Nguyen Doan, Mojtaba Vaezi, Wonjae Shin, H. Vincent Poor, Hyundong Shin, Tony Q. S. Quek
2019 arXiv   pre-print
The second one is based on the deep reinforcement learning method that allows all users to share the full bandwidth.  ...  In order to optimize users' quality of service and, concurrently, ensure the fairness among users, the probability that all users can decode the desired signals is maximized.  ...  Deep-Reinforcement-Learning-Based Scheme 1) Problem Re-formulation: Before introducing a new form of problem formulation such that a learning method can be applied, we would like to mention the concept  ... 
arXiv:1909.11074v1 fatcat:frxzl3vt3bdbvpaen3x4iufpuq

Optimized Computation Offloading Performance in Virtual Edge Computing Systems via Deep Reinforcement Learning

Xianfu Chen, Honggang Zhang, Celimuge Wu, Shiwen Mao, Yusheng Ji, Mehdi Bennis
2018 IEEE Internet of Things Journal  
To break the curse of high dimensionality in state space, we first propose a double deep Q-network (DQN) based strategic computation offloading algorithm to learn the optimal policy without knowing a priori  ...  Numerical experiments show that our proposed learning algorithms achieve a significant improvement in computation offloading performance compared with the baseline policies.  ...  Fig. 3 . 3 Deep SARSA reinforcement learning (Deep-SARL) based stochastic computation offloading in a mobile-edge computing system.  ... 
doi:10.1109/jiot.2018.2876279 fatcat:6gsargvoffgjpkqisuot6yp42e

Deep reinforcement learning and its application in autonomous fitting optimization for attack areas of UCAVs

Li Yue, Qiu Xiaohui, Liu Xiaodong, Xia Qunli
2020 Journal of Systems Engineering and Electronics  
To solve the problem, this paper proposes a new deep deterministic policy gradient (DDPG) strategy based on deep reinforcement learning for the attack area fitting of UCAVs in the future battlefield.  ...  We can obtain the optimal values of attack areas in real time during the whole flight with the well-trained deep network.  ...  Deep learning (DL) is widely used in UCAV data processing [20 -23] . Reinforcement learning (RL) is widely used in UCAV autonomous decision making [24 -26] .  ... 
doi:10.23919/jsee.2020.000048 fatcat:3lxk5wk7zzhh7nm2jyht2p3eii

Augmented Q Imitation Learning (AQIL) [article]

Xiao Lei Zhang, Anish Agarwal
2020 arXiv   pre-print
In imitation learning the machine learns by mimicking the behavior of an expert system whereas in reinforcement learning the machine learns via direct environment feedback.  ...  Traditional deep reinforcement learning takes a significant time before the machine starts to converge to an optimal policy.  ...  In both Q-imitation-learning and deep reinforcement learning, we used the Q-learning methodology for training.  ... 
arXiv:2004.00993v2 fatcat:nwtzjuijwfgqfd3paiu34iy7my

A DPDK-Based Acceleration Method for Experience Sampling of Distributed Reinforcement Learning [article]

Masaki Furukawa, Hiroki Matsutani
2022 arXiv   pre-print
In distributed reinforcement learning, Actor nodes acquire experiences by interacting with a given environment and a Learner node optimizes their DQN model.  ...  A computing cluster that interconnects multiple compute nodes is used to accelerate distributed reinforcement learning based on DQN (Deep Q-Network).  ...  A distributed deep reinforcement learning system is implemented and analyzed in terms of network overheads in Section 3, and the proposed network optimizations are applied to the system in Section 4.  ... 
arXiv:2110.13506v2 fatcat:s34wxpnohjdypmcokxfroi22mi

Special Issue on Deep Reinforcement Learning for Emerging IoT Systems

Jia Hu, Peng Liu, Hong Liu, Obinna Anya, Yan Zhang
2020 IEEE Internet of Things Journal  
Jia Hu received the B.Eng. and M.Eng. degrees in electronic engineering from the Huazhong  ...  Peng Liu received the B.S. and M.S. degrees in computer science and technology from Hangzhou Dianzi University, Hangzhou, China, in 2001 and 2004, respectively, and  ...  The article titled "A deep-reinforcement-learning-based recommender system for occupant-driven energy optimization in commercial buildings" presents recEnergy, a DRL-based recommender system to reduce  ... 
doi:10.1109/jiot.2020.2998256 fatcat:rct75tsesbh7lkjmhuyogfi4ym

Deep reinforcement learning for optical systems: A case study of mode-locked lasers [article]

Chang Sun, Eurika Kaiser, Steven L. Brunton, J. Nathan Kutz
2020 arXiv   pre-print
We demonstrate that deep reinforcement learning (deep RL) provides a highly effective strategy for the control and self-tuning of optical systems.  ...  This allows the optical system to recognize bi-stable structures and navigate, via trajectory planning, to optimally performing solutions, the first such algorithm demonstrated to do so in optical systems  ...  Steps in each episode DISCUSSION Deep reinforcement learning is a learning paradigm that integrates the power of reinforcement learning and deep neural networks.  ... 
arXiv:2006.05579v1 fatcat:xqxghch6rre55dkgnf6wfjicjq

A Cognitive Relay Network Throughput Optimization Algorithm Based on Deep Reinforcement Learning

Shaojiang Liu, Kejing Hu, Weichuan Ni, Zhiming Xu, Feng Wang, Zhiping Wan
2019 Wireless Communications and Mobile Computing  
In order to improve the throughput of cognitive relay network and optimize system utility, this paper proposes a cognitive relay network throughput optimization algorithm based on deep reinforcement learning  ...  Then, the maximum utility optimization strategy based on deep reinforcement learning is proposed to maximize the system utility revenue by selecting the optimal behavior.  ...  system throughput and power optimization problems, a system utility function is proposed to maximize the benefits by selecting the optimal behavior by deep reinforcement learning algorithm; (4) in the  ... 
doi:10.1155/2019/2731485 fatcat:6mijh75lnrfadg227rfvx5kvqa
« Previous Showing results 1 — 15 out of 110,805 results