Filters








72,856 Hits in 4.9 sec

Self-Optimizing Memory Controllers: A Reinforcement Learning Approach

Engin Ipek, Onur Mutlu, José F. Martínez, Rich Caruana
2008 2008 International Symposium on Computer Architecture  
We propose a new, self-optimizing memory controller design that operates using the principles of reinforcement learning (RL) to overcome these limitations.  ...  Key idea: We propose to design the memory controller as an RL agent whose goal is to learn automatically an optimal memory scheduling policy via interaction with the rest of the system.  ...  [43] explore a reinforcement learning approach to make autonomic resource allocation decisions in data centers. They focus on assigning processors and memory to applications.  ... 
doi:10.1109/isca.2008.21 dblp:conf/isca/IpekMMC08 fatcat:ct2pofqkrfaxlbtxvnp3corlam

Self-Optimizing Memory Controllers

Engin Ipek, Onur Mutlu, José F. Martínez, Rich Caruana
2008 SIGARCH Computer Architecture News  
We propose a new, self-optimizing memory controller design that operates using the principles of reinforcement learning (RL) to overcome these limitations.  ...  Key idea: We propose to design the memory controller as an RL agent whose goal is to learn automatically an optimal memory scheduling policy via interaction with the rest of the system.  ...  [43] explore a reinforcement learning approach to make autonomic resource allocation decisions in data centers. They focus on assigning processors and memory to applications.  ... 
doi:10.1145/1394608.1382172 fatcat:rqvs4uqidfdqhkjfzpbzpvtdqm

Safe RAN control: A Symbolic Reinforcement Learning Approach [article]

Alexandros Nikou, Anusha Mujumdar, Vaishnavi Sundararajan, Marin Orlic, Aneta Vulgarakis Feljan
2022 arXiv   pre-print
In this paper, we present a Symbolic Reinforcement Learning (SRL) based architecture for safety control of Radio Access Network (RAN) applications.  ...  Recent research has focused on performing this RET optimization by employing Reinforcement Learning (RL) strategies due to the fact that they have self-learning capabilities to adapt in uncertain environments  ...  Reinforcement Learning based RAN control Consider an area covered by R Radio Base Stations (RBS) with C cells that serve a set of U UEs uniformly distributed in the area (see Fig. 1 ).  ... 
arXiv:2106.01977v2 fatcat:ozwmrtmrfjcsnb7c7dkn5x2fpu

A Deep Reinforcement Learning Approach for Fair Traffic Signal Control [article]

Majid Raeis, Alberto Leon-Garcia
2021 arXiv   pre-print
In recent years, traffic control methods based on deep reinforcement learning (DRL) have gained attention due to their ability to exploit real-time traffic data, which is often poorly used by the traditional  ...  Furthermore, we propose two DRL-based traffic signal control methods for implementing these fairness notions, that can achieve a high throughput as well.  ...  METHODOLOGY In this section, we present our method for providing fairness using a reinforcement learning approach. The environment is an intersection as shown in Fig. 1 .  ... 
arXiv:2107.10146v1 fatcat:khuoxti4vrd6ldpmn3pzemgvna

Towards Self‐Driving Processes: A Deep Reinforcement Learning Approach to Control

Steven Spielberg, Aditya Tulsyan, Nathan P. Lawrence, Philip D Loewen, R. Bhushan Gopaluni
2019 AIChE Journal  
In this paper, we exploit recent developments in reinforcement learning and deep learning to develop a novel adaptive, model-free controller for general discrete-time processes.  ...  The DRL controller we propose is a data-based controller that learns the control policy in real time by merely interacting with the process.  ...  Algorithm 5 Deep Reinforcement Learning Controller1: Output: Optimal policy µ(s, W a ) 2: Initialize: W a , W c to random initial weights 3: Initialize: W a ← W a and W c ← W c 4: Initialize: Replay memory  ... 
doi:10.1002/aic.16689 fatcat:2po2quoyfrakpmqcfcsfp6tare

Automated Vehicle Control at Freeway Lane-drops: a Deep Reinforcement Learning Approach

Salaheldeen M. S. Seliman, Adel W. Sadek, Qing He
2020 Journal of Big Data Analytics in Transportation  
This study develops an optimal, real-time and adaptive control algorithm for helping a Connected and Automated Vehicle (CAV), navigate a freeway lane-drop site (e.g. work zones).  ...  The proposed traffic control strategy is based on the Deep Q-Network (DQN) Reinforcement Learning (RL) algorithm, and is designed to determine the driving speed and lane-change maneuvers that would enable  ...  The following sub-sections will describe our approach in more detail. Reinforcement Learning (RL) Reinforcement learning, a machine learning method that stemmed out from the field of psychology.  ... 
doi:10.1007/s42421-020-00021-0 fatcat:eia3aqy2bvampiloondygil5pu

Tiyuntsong: A Self-Play Reinforcement Learning Approach for ABR Video Streaming [article]

Tianchi Huang, Xin Yao, Chenglei Wu, Rui-Xiao Zhang, Zhangyuan Pang, Lifeng Sun
2019 arXiv   pre-print
In this paper, we propose Tiyuntsong, a self-play reinforcement learning approach with generative adversarial network (GAN)-based method for ABR video streaming.  ...  Existing reinforcement learning (RL)-based adaptive bitrate (ABR) approaches outperform the previous fixed control rules based methods by improving the Quality of Experience (QoE) score, as the QoE metric  ...  In this paper, we prove that using self-play reinforcement learning will learn the strategy by itself if you can tell the agent who is better. E.2.  ... 
arXiv:1811.06166v3 fatcat:eoghii3bcfeobdqjfjud3sgbju

Reinforcement-Learning-Based Vibration Control for a Vehicle Semi-Active Suspension System via the PPO Approach

Shi-Yuan Han, Tong Liang
2022 Applied Sciences  
The main contribution of this paper is the proposal of a PPO-based vibration control strategy for a vehicle semi-active suspension system, in which the designed reward function realizes the dynamic adjustment  ...  The simulation results showed that the body acceleration was reduced by 46.93% under the continuously changing road, which proved that the control strategy could effectively improve the performance of  ...  PPO Algorithm Network Model The proximal policy optimization (PPO) algorithm is an actor-critic-based reinforcement learning algorithm, which combines a value-based approach and a policy-based approach  ... 
doi:10.3390/app12063078 fatcat:stbodw72wzbiri7d224i5czmle

Load Balancing for Ultra-Dense Networks: A Deep Reinforcement Learning Based Approach [article]

Yue Xu and Wenjun Xu and Zhi Wang and Jiaru Lin and Shuguang Cui
2019 arXiv   pre-print
In this paper, we propose a deep reinforcement learning (DRL) based mobility load balancing (MLB) algorithm along with a two-layer architecture to solve the large-scale load balancing problem for ultra-dense  ...  First, this work proposes a two-layer architecture to solve the large-scale load balancing problem in a self-organized manner.  ...  Preliminaries on Reinforcement Learning 1) Reinforcement Learning: In this paper, we formulate the load balancing problem with RL.  ... 
arXiv:1906.00767v3 fatcat:eqyjmlhnebhnbkrq6tlrsdwjwy

A reinforcement learning approach to dynamic resource allocation

David Vengerov
2007 Engineering applications of artificial intelligence  
A specific architecture (DRA-FRL) is presented, which uses the emerging methodology of reinforcement learning in conjunction with fuzzy rulebases to achieve the desired objective.  ...  An implementation of the DRA-FRL architecture in Solaris 10 demonstrated a robust performance improvement in the problem of dynamically migrating CPUs and memory blocks between three resource partitions  ...  Future work A possible approach to avoiding the "arms race" in the context of multi-agent reinforcement learning was presented in [10] .  ... 
doi:10.1016/j.engappai.2006.06.019 fatcat:mqimb6yrfbh5td4sfcn55otko4

Control Strategy of a Hybrid Renewable Energy System Based on Reinforcement Learning Approach for an Isolated Microgrid

Phan, Lai
2019 Applied Sciences  
Recent developments and achievements in the fields of machine learning (ML) and reinforcement learning (RL) have led to new challenges and opportunities for HRES development.  ...  According to the assessment of EMS and MPPT control of HRES, it can be concluded that RL is one of the most emerging optimal control solutions.  ...  The reinforcement learning approach to solve the MPPT problem aims to learn the system behavior based on the PV source response.  ... 
doi:10.3390/app9194001 fatcat:mk7cqz35kjhu5isoxxh2xzovky

A Meta-MDP Approach to Exploration for Lifelong Reinforcement Learning [article]

Francisco M. Garcia, Philip S. Thomas
2019 arXiv   pre-print
In this paper we consider the problem of how a reinforcement learning agent that is tasked with solving a sequence of reinforcement learning problems (a sequence of Markov decision processes) can use knowledge  ...  We show that the search for an optimal exploration strategy can be formulated as a reinforcement learning problem itself and demonstrate that such strategy can leverage patterns found in the structure  ...  This difference of timescales distinguishes our approach from previous meta-MDP methods for optimizing components of reinforcement learning algorithms, [25, 11, 26, 10, 4] .  ... 
arXiv:1902.00843v1 fatcat:a3r6l73fnnelxen3grpzucye2a

A Reinforcement Learning Approach to the Orienteering Problem with Time Windows [article]

Ricardo Gama, Hugo L. Fernandes
2021 arXiv   pre-print
A neural network allows learning solutions using reinforcement learning or supervised learning, depending on the available data.  ...  Once a model-region is trained, it can infer a solution for a particular tourist using beam search. We based the assessment of our approach on several existing benchmark OPTW instances.  ...  The exploration strategy in a reinforcement learning approach can also have a big impact on performance and learning speed.  ... 
arXiv:2011.03647v2 fatcat:5ydvwopyzfcwnpsrqkx7izsblu

Resource Allocation in Uplink NOMA-IoT Networks: A Reinforcement-Learning Approach [article]

Waleed Ahsan, Wenqiang Yi, Zhijin Qin, Yuanwei Liu, Arumugam Nallanathan
2021 arXiv   pre-print
To maximise the average performance of sum rates, this work designs an efficient optimization approach based on two reinforcement learning algorithms, namely deep reinforcement learning (DRL) and SARSA-learning  ...  Numerical results show that: 1) Compared with the optimal benchmark scheme, the proposed DRL and SARSA-learning algorithms have lower complexity with acceptable accuracy and 2) NOMA-enabled IoT networks  ...  For large-scale NOMA-IoT networks, an intelligent reinforcement learning (RL) algorithm becomes a promising approach to find the optimal long-term resource allocation strategy.  ... 
arXiv:2007.08350v2 fatcat:aqukni5yyjf45hsfyenchp2kyi

Energy-Efficient UAV Movement Control for Fair Communication Coverage: A Deep Reinforcement Learning Approach

Ibrahim A. Nemer, Tarek R. Sheltami, Slim Belhaiza, Ashraf S. Mahmoud
2022 Sensors  
Then, we merge SBG-AC with an actor–critic algorithm to assure the convergence of the model, to control each UAV in a distributed way, and to have learning capabilities in case of dynamic environments.  ...  In this paper, we introduced a novel distributed control solution to place a group of UAVs in the candidate area in order to improve the coverage score with minimum energy consumption and a high fairness  ...  Furthermore, in [36] , Liu et al. suggested a distributed deep reinforcement learning approach for controlling the UAV in a decentralized way.  ... 
doi:10.3390/s22051919 pmid:35271067 pmcid:PMC8915037 fatcat:hfrhzxxgxfhyzjam37y7ydfify
« Previous Showing results 1 — 15 out of 72,856 results