45 Hits in 4.1 sec

Congested Bandits: Optimal Routing via Short-term Resets

Pranjal Awasthi, Kush Bhatia, Sreenivas Gollapudi, Kostas Kollias
2022 International Conference on Machine Learning  
We study the congestion aware formulation in the multi-armed bandit (MAB) setup and in the contextual bandit setup with linear rewards.  ...  For traffic routing platforms, the choice of which route to recommend to a user depends on the congestion on these routes -indeed, an individual's utility depends on the number of people using the recommended  ...  We propose and study the congested bandits model with short term resets under a variety of settings and design no-regret algorithms.  ... 
dblp:conf/icml/AwasthiBGK22 fatcat:6lwb7y4f6zetzigdfledissu7m

Multi-Armed Bandit in Action: Optimizing Performance in Dynamic Hybrid Networks

Sebastien Henri, Christina Vlachou, Patrick Thiran
2018 IEEE/ACM Transactions on Networking  
This is, to the best of our knowledge, the first implementation on a real test bed of multi-armed-bandit strategies in the context of routing.  ...  We employ the multi-armed-bandit framework and prove that HyMAB achieves optimal throughput under a static scenario.  ...  EMPoWER [2] is a multipath congestion-control and routing system for hybrid networks with shared-medium technologies. It optimizes throughput and controls the congestion on a single multipath.  ... 
doi:10.1109/tnet.2018.2856302 fatcat:nvhtf6bznrga3j6vlfimq3rwwy

Practical Adversarial Combinatorial Bandit Algorithm via Compression of Decision Sets [article]

Shinsaku Sakaue, Masakazu Ishihata, Shin-ichi Minato
2017 arXiv   pre-print
We consider the adversarial combinatorial multi-armed bandit (CMAB) problem, whose decision set can be exponentially large with respect to the number of given arms.  ...  Experimental results show that our algorithm is applicable to various large adversarial CMAB instances including adaptive routing problems on real-world networks.  ...  This setting is a stochastic CMAB with distributions Ber(µ t,i ) in the short run, but the adversary secretly reset µ t with probability 0.1 in each round to foil the player.  ... 
arXiv:1707.08300v1 fatcat:tk2fbknzorgtnazqs7i56ft2ca

Efficient Bandit Combinatorial Optimization Algorithm with Zero-suppressed Binary Decision Diagrams

Shinsaku Sakaue, Masakazu Ishihata, Shin-ichi Minato
2018 International Conference on Artificial Intelligence and Statistics  
We consider bandit combinatorial optimization (BCO) problems. A BCO instance generally has a huge set of all feasible solutions, which we call the action set.  ...  Experiments show that our algorithm is applicable to various large BCO instances including adaptive routing problems on real-world networks.  ...  In the short run, this setting is a stochastic OCO problem where each `t,i is drawn from Ber(µ t,i ), but the adversary secretly resets µ t with probability 0.1 in each round to foil the player. 2 (e )  ... 
dblp:conf/aistats/SakaueIM18 fatcat:fjq52w63ozeh3ciffzmhvu46oy

SDMob: SDN-Based Mobility Management for IoT Networks

Iliar Rabet, Shunmunga Priyan Selvaraju, Hossein Fotouhi, Mário Alves, Maryam Vahabi, Ali Balador, Mats Björkman
2022 Journal of Sensor and Actuator Networks  
To keep the connectivity of the constrained nodes upon topological changes, it is of vital importance to enhance the standard protocol stack, including the Routing Protocol for Lossy Low-power Networks  ...  Through analytical modeling and simulations, we show that SDMob outperforms the baseline RPL and the state-of-the-art ARMOR in terms of packet delivery ratio and end-to-end delay, with an adjustable and  ...  This allows BRPL to utilize sub-optimal routes when the optimal route is congested. In mRPL [13] , MN operates in two phases of data transmission and discovery.  ... 
doi:10.3390/jsan11010008 fatcat:ntf4muefvre4vkjo3ekyqu5u5i

Assessing Routing Strategies for Cognitive Radio Sensor Networks

Suleiman Zubair, Norsheila Fisal, Yakubu Baguda, Kashif Saleem
2013 Sensors  
With the successful implementation of DSA via cognitive radio (CR), other advantages are exploited by the WSN.  ...  This practice, in effect, can also enable the coexistence of various WSNs deployed in a spatially overlapping area in terms of communication and resource utilization.  ...  Multipath routing is modeled as a restless bandit stochastic process optimization problem that allows secondary users to select routes considering the dynamic occupancy of a licensed spectrum and energy  ... 
doi:10.3390/s131013005 pmid:24077319 pmcid:PMC3859047 fatcat:xdle3amrnndxrmvjgnngfg6c5i

Reinforcement Learning-Enabled Cross-Layer Optimization for Low-Power and Lossy Networks under Heterogeneous Traffic Patterns

Arslan Musaddiq, Zulqar Nain, Yazdan Ahmad Qadri, Rashid Ali, Sung Won Kim
2020 Sensors  
Similarly, the network layer uses a ranking mechanism to route the packets.  ...  The next generation of the Internet of Things (IoT) networks is expected to handle a massive scale of sensor deployment with radically heterogeneous traffic applications, which leads to a congested network  ...  long-term rewards.  ... 
doi:10.3390/s20154158 pmid:32722645 fatcat:eimk5zm3kvbdfaqq7mci452lw4

Reinforcement Learning for Ridesharing: An Extended Survey [article]

Zhiwei Qin, Hongtu Zhu, Jieping Ye
2022 arXiv   pre-print
Papers on the topics of rideshare matching, vehicle repositioning, ride-pooling, routing, and dynamic pricing are covered.  ...  In this paper, we present a comprehensive, in-depth survey of the literature on reinforcement learning approaches to decision optimization problems in a typical ridesharing system.  ...  into the bandit rewards.  ... 
arXiv:2105.01099v7 fatcat:36x75go5b5hhdmdigbuzdzundu

Cache Optimization Models and Algorithms [article]

Georgios Paschos and George Iosifidis and Giuseppe Caire
2019 arXiv   pre-print
Therefore, it is crucial to design these systems in an optimal fashion, ensuring the maximum possible performance and economic benefits from their deployment.  ...  To this end, this article presents a collection of detailed models and algorithms, which are synthesized to build a powerful analytical framework for caching optimization.  ...  For example, the closest cache to a node might be reachable only via highly congested links.  ... 
arXiv:1912.12339v1 fatcat:tofgtzkzx5cudpepyhodwbvnnm

RL-based Resource Allocation in mmWave 5G IAB Networks

Bibo Zhang, Francesco Devoti, Ilario Filippini
2020 2020 Mediterranean Communication and Computer Networking Conference (MedComNet)  
For this reason, traditional optimization techniques do not provide the best performance in these conditions.  ...  Routing and scheduling in wireless multi-hop networks are typically carried out via optimization techniques considering all available links [2] [3] [4] .  ...  Accordingly, the optimization objective (1) serves as the long-term expected return RL maximizes.  ... 
doi:10.1109/medcomnet49392.2020.9191546 fatcat:ulgsfibszfhotdbm4om3esfopq

Optimal data scheduling of mobile clients serviced using beamforming antennas

Daniel T. Bennett, Timothy X Brown
2012 MILCOM 2012 - 2012 IEEE Military Communications Conference  
short term.  ...  So, it achieves better contention control and short-term fairness as well as decoupling contention control from handling packet losses.  ... 
doi:10.1109/milcom.2012.6415855 dblp:conf/milcom/BennettB12 fatcat:yy62tleh4bcapjdkdbx7k5uvu4

Initial Access & Beam Alignment for mmWave and Terahertz Communications

Wissal Attaoui, Khadija Bouraqia, Essaid Sabir
2022 IEEE Access  
INDEX TERMS 5G/6G, mmWave, terahertz, beamforming, initial access, beam alignment, beam steering, reconfigurable intelligent surface.  ...  beamforming via antenna arrays.  ...  Simulation results prove the efficiency of the sub-optimal solution in terms of achievable sum rate compared to conventional OMA mmWave.  ... 
doi:10.1109/access.2022.3161951 fatcat:wgwrvxyz7relpb3afo4rq3nbli

Age of Information: An Introduction and Survey [article]

Roy D. Yates, Yin Sun, D. Richard Brown III, Sanjit K. Kaul, Eytan Modiano, Sennur Ulukus
2020 arXiv   pre-print
The paper concludes with a review of efforts to employ age optimization in cyberphysical applications.  ...  In particular, we describe the current state of the art in the design and optimization of low-latency cyberphysical systems and applications in which sources send time-stamped status updates to interested  ...  Whittle's Index policy is the optimal solution to a relaxation of the Restless Multi-Armed Bandit (RMAB) problem.  ... 
arXiv:2007.08564v1 fatcat:l7ctda3ukfge5hqtbk3ez7pjia

Applications of Deep Reinforcement Learning in Communications and Networking: A Survey

Nguyen Cong Luong, Dinh Thai Hoang, Shimin Gong, Dusit Niyato, Ping Wang, Ying-Chang Liang, Dong In Kim
2019 IEEE Communications Surveys and Tutorials  
Furthermore, we present applications of deep reinforcement learning for traffic routing, resource sharing, and data collection.  ...  However, in complex and large-scale networks, the state and action spaces are usually large, and the reinforcement learning may not be able to find the optimal policy in reasonable time.  ...  Thus, Long Short-Term Memory (LSTM) is often used in RNNs to address this issue.  ... 
doi:10.1109/comst.2019.2916583 fatcat:5owsswhhrbctnirdtxre6mhv24

Deep Reinforcement Learning [article]

Yuxi Li
2018 arXiv   pre-print
StarCraft features many possible actions, complex interactions between players, short term tactics and long term strategies, etc.  ...  A hypothesis is that we learn on two timescales: learning on specific examples in short term, and learning abstract skills or rules over long term.  ...  Speech-centric information processing: An optimization-oriented approach.  ... 
arXiv:1810.06339v1 fatcat:kp7atz5pdbeqta352e6b3nmuhy
« Previous Showing results 1 — 15 out of 45 results