Filters








13,948 Hits in 6.4 sec

A Reinforcement Learning - Great-Deluge Hyper-Heuristic for Examination Timetabling

Ender Özcan, Mustafa Misir, Gabriela Ochoa, Edmund K. Burke
2010 International Journal of Applied Metaheuristic Computing  
Hyper-heuristics can be identified as methodologies that search the space generated by a finite set of low level heuristics for solving search problems.  ...  An iterative hyper-heuristic framework can be thought of as requiring a single candidate solution and multiple perturbation low level heuristics.  ...  Plot of utility value for each low level heuristic and quality versus iteration on sta83 I using the reinforcement learning -great deluge hyper-heuristic based on (a) subtractive, (b) divisional and (c  ... 
doi:10.4018/jamc.2010102603 fatcat:6mpraj2csvg7pplao77jephkgu

Examination timetabling using late acceptance hyper-heuristics

Ender Ozcan, Yuri Bykov, Murat Birben, Edmund K. Burke
2009 2009 IEEE Congress on Evolutionary Computation  
One of the hyperheuristic frameworks is based on a single point search containing two main stages: heuristic selection and move acceptance.  ...  Most of the existing move acceptance methods compare a new solution, generated after applying a heuristic, against a current solution in order to decide whether to reject it or replace the current one.  ...  Learning mechanisms based on reinforcement learning or statistical analyses do not function well in combination with the late acceptance strategy.  ... 
doi:10.1109/cec.2009.4983054 dblp:conf/cec/OzcanBBB09 fatcat:ffcorbkfavdg3p6mg5uy6vvcyu

Multi-Tenant Provisioning for Quantum Key Distribution Networks with Heuristics and Reinforcement Learning: A Comparative Study

Yuan Cao, Yongli Zhao, Jun Li, Rui Lin, Jie Zhang, Jiajia Chen
2020 IEEE Transactions on Network and Service Management  
To realize efficient On-MTP, we perform a comparative study of heuristics and reinforcement learning (RL) based On-MTP solutions, where three heuristics (i.e., random, fit, and best-fit based On-MTP algorithms  ...  Index Terms-Quantum key distribution networks, online multi-tenant provisioning, heuristics, reinforcement learning.  ...  Accordingly, the RL might provide a viable alternative to heuristics for On-MTP, which performs the task of learning how an agent should take a series of actions in an environment in order to maximize  ... 
doi:10.1109/tnsm.2020.2964003 fatcat:xkuhmbrh4bd7fiqqdaxmeaf57m

Deep Reinforcement Learning for URLLC data management on top of scheduled eMBB traffic [article]

Fabio Saggese, Luca Pasqualini, Marco Moretti, Andrea Abrardo
2021 arXiv   pre-print
In this paper we propose a deep reinforcement learning (DRL) algorithm to slice the available physical layer resources between ultra-reliable low-latency communications (URLLC) and enhanced Mobile BroadBand  ...  Specifically, in our setting the time-frequency resource grid is fully occupied by eMBB traffic and we train the DRL agent to employ proximal policy optimization (PPO), a state-of-the-art DRL algorithm  ...  Reinforcement Learning Reinforcement Learning (RL) is usually employed to solve a Markov Decision Process (MDP) defined over a real world task.  ... 
arXiv:2103.01801v1 fatcat:ghvkfwfarzhxxea3vdeekzss4u

A framework for meta-level control in multi-agent systems

Anita Raja, Victor Lesser
2007 Autonomous Agents and Multi-Agent Systems  
This is the meta-level control problem for agents operating in resource-bounded multi-agent environments.  ...  This abstraction concisely captures critical information necessary for decision making while bounding the cost of meta-level control and is appropriate for use in automatically learning the meta-level  ...  Acknowledgments We would like to thank Professor Shlomo Zilberstein for his help in constructing the model described in Section 2.1 and Professor Andy Barto for his valuable comments on the Reinforcement  ... 
doi:10.1007/s10458-006-9008-z fatcat:lfukqobaqzcephqkxvwzndf3l4

Cloud Task Scheduling Based on Proximal Policy Optimization Algorithm for Lowering Energy Consumption of Data Center

2022 KSII Transactions on Internet and Information Systems  
As a part of cloud computing technology, algorithms for cloud task scheduling place an important influence on the area of cloud computing in data centers.  ...  In our earlier work, we proposed DeepEnergyJS, which was designed based on the original version of the policy gradient and reinforcement learning algorithm.  ...  Q-learning [5] is a classic algorithm for reinforcement learning.  ... 
doi:10.3837/tiis.2022.06.006 fatcat:xhxte7kjazdf7moisktk6xwcpi

A greedy gradient-simulated annealing selection hyper-heuristic

Murat Kalender, Ahmed Kheiri, Ender Özcan, Edmund K. Burke
2013 Soft Computing - A Fusion of Foundations, Methodologies and Applications  
In most of the previous applications of reinforcement learning in hyper-heuristics, a utility value is increased as a reward mechanism and decreased for punishment (Nareyek, 2004; Bai et al, 2007a) .  ...  Burke et al (2003) used a combination of tabu search and reinforcement learning scheme as a heuristic selector and tested their hyper-heuristic over different timetabling problems.  ... 
doi:10.1007/s00500-013-1096-5 fatcat:yklgq6fclvbnxfiyyksscnja54

Solving high school timetabling problems worldwide using selection hyper-heuristics

Leena N. Ahmed, Ender Özcan, Ahmed Kheiri
2015 Expert systems with applications  
Solving a high school timetabling problem requires scheduling of resources and events into time slots subject to a set of constraints.  ...  In this study, we evaluate the performance of a range of selection hyper-heuristics combining different reusable components for high school timetabling.  ...  The reinforcement learning based adaptive heuristic selection method CF performs the worst when combined with the adaptive move acceptance method GD for high school timetabling, while RP, a heuristic selection  ... 
doi:10.1016/j.eswa.2015.02.059 fatcat:yvqnom23vzd3ho3y2vri62safu

Learning heuristics for basic block instruction scheduling

Abid M. Malik, Tyrel Russell, Michael Chase, Peter van Beek
2007 Journal of Heuristics  
The heuristic is usually hand-crafted, a potentially time-consuming process. In contrast, we present a study on automatically learning good heuristics using techniques from machine learning.  ...  A fundamental problem that arises in instruction scheduling is to find a minimum length schedule for a basic block-a straight-line sequence of code with a single entry point and a single exit point-subject  ...  Acknowledgements This research was supported by an IBM Center for Advanced Studies (CAS) Fellowship, an NSERC Postgraduate Scholarship, and an NSERC CRD Grant.  ... 
doi:10.1007/s10732-007-9051-1 fatcat:yzougfgpr5dhvfsct7nryuklka

LEARNING TO SUPPORT CONSTRAINT PROGRAMMERS

Susan L. Epstein, Eugene C. Freuder, Richard J. Wallace
2005 Computational intelligence  
The program harnesses a cognitively oriented architecture-FOr the Right Reasons (FORR) to manage search heuristics and to learn new ones.  ...  It currently serves both as a learner and as a test bed for the constraint community.  ...  A simple constraint problem and its underlying constraint graph. Labels on the edges give acceptable values for the variables in alphabetical order.  ... 
doi:10.1111/j.1467-8640.2005.00277.x fatcat:mm5t3bh74rhfrghgfkjcuy2aby

Reinforcement Learning for Adaptive Routing [article]

Leonid Peshkin, Virginia Savova
2007 arXiv   pre-print
Reinforcement learning means learning a policy--a mapping of observations into actions--based on feedback from the environment.  ...  We present an application of gradient ascent algorithm for reinforcement learning to a complex domain of packet routing in network communication and compare the performance of this algorithm to other routing  ...  Algorithmic details Williams introduced the notion of policy search via gradient ascent for reinforcement learning in his reinforce algorithm [18] , [19] , which was generalized to a broader class of  ... 
arXiv:cs/0703138v1 fatcat:dyks5xbvonegre6zgtqhwebcgm

HyMER: A Hybrid Machine Learning Framework for Energy Efficient Routing in SDN [article]

Beakal Gizachew Assefa, Oznur Ozkasap
2020 arXiv   pre-print
To the best of our knowledge, HyMER is the first that utilizes a hybrid machine learning solution with supervised and reinforcement learning components for energy efficiency and network performance in  ...  Addressing the significance of energy efficiency in networks, we propose a novel hybrid machine learning-based framework named HyMER that combines the capabilities of SDN and machine learning for traffic-aware  ...  A very preliminary version of this work was presented in [30] .  ... 
arXiv:1909.08074v2 fatcat:7pxl542pdfh73nnfqbhknkuxoe

Throughput-Aware Cooperative Reinforcement Learning for Adaptive Resource Allocation in Device-to-Device Communication

Muhidul Khan, Muhammad Alam, Yannick Moullec, Elias Yaacoub
2017 Future Internet  
In this paper, we propose a cooperative reinforcement learning algorithm for adaptive resource allocation, which contributes to improving system throughput.  ...  To illustrate the problem, Figure 1 shows a basic single cell scenario with one Cellular user (CU), two D2D pairs and one base station having two resource blocks operating in an underlay mode.  ...  Elias Yaacoub provided useful suggestions based on to improve the system model and basic components of the method.  ... 
doi:10.3390/fi9040072 fatcat:xnnpbeq2ffcxblk75pr3b77a2e

Layer-compensated Pruning for Resource-constrained Convolutional Neural Networks [article]

Ting-Wu Chin, Cha Zhang, Diana Marculescu
2018 arXiv   pre-print
Resource-efficient convolution neural networks enable not only the intelligence on edge devices but also opportunities in system-level optimization such as scheduling.  ...  Our framework entails a novel algorithm, dubbed layer-compensated pruning, where meta-learning is involved to determine better solutions.  ...  ., 2018b) approach this problem with reinforcement learning to learn an agent for deciding how many filters to prune for each layer given a resource constraint.  ... 
arXiv:1810.00518v2 fatcat:yk7l2fflxbejfoshmxqumzmzfm

Resource allocation optimization using artificial intelligence methods in various computing paradigms: A Review [article]

Javad Hassannataj Joloudari, Roohallah Alizadehsani, Issa Nodehi, Sanaz Mojrian, Fatemeh Fazl, Sahar Khanjani Shirkharkolaie, H M Dipu Kabir, Ru-San Tan, U Rajendra Acharya
2022 arXiv   pre-print
This paper presents a comprehensive literature review on the application of artificial intelligence (AI) methods such as deep learning (DL) and machine learning (ML) for resource allocation optimization  ...  The review ends with a discussion on open research directions and a conclusion.  ...  Edge computing Categorized reinforcement learning and heuristic learning methods for public safety communications on 5G networks ML Atman [24] Fog computing Examined categories of resource management  ... 
arXiv:2203.12315v1 fatcat:43mouwxwene6xllnw3gsmdh6hy
« Previous Showing results 1 — 15 out of 13,948 results