How network monitoring and reinforcement learning can improve tcp fairness in wireless multi-hop networks

Arianpoo, Nasim; Leung, Victor C. M.

doi:10.1186/s13638-016-0773-3

Cited by 10 publications

(8 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Leung [6] propose a distributed mechanism that observes the unfairness in resource allocation and then tunes the TCP parameters accordingly. In their solution, each TCP source models the state of the multi-hop network as an MDP, and then they use the Q-learning algorithm to monitor and learn the transition matrix of the proposed MDP.…”

Section: Implementations Of Fair-rl For Non-societal Fairnessmentioning

confidence: 99%

Survey on Fair Reinforcement Learning: Theory and Practice

Gajane¹,

Saxena²,

Tavakol³

et al. 2022

Preprint

View full text Add to dashboard Cite

Fairness-aware learning aims at satisfying various fairness constraints in addition to the usual performance criteria via data-driven machine learning techniques. Most of the research in fairness-aware learning employs the setting of fair-supervised learning. However, many dynamic real-world applications can be better modeled using sequential decision-making problems and fair reinforcement learning provides a more suitable alternative for addressing these problems. In this article, we provide an extensive overview of fairness approaches that have been implemented via a reinforcement learning (RL) framework. We discuss various practical applications in which RL methods have been applied to achieve a fair solution with high accuracy. We further include various facets of the theory of fair reinforcement learning, organizing them into single-agent RL, multi-agent RL, long-term fairness via RL, and offline learning. Moreover, we highlight a few major issues to explore in order to advance the field of fair-RL, namely -i) correcting societal biases, ii) feasibility of group fairness or individual fairness, and iii) explainability in RL. Our work is beneficial for both researchers and practitioners as we discuss articles providing mathematical guarantees as well as articles with empirical studies on real-world problems.CCS Concepts: • Computing methodologies → Machine learning approaches.

show abstract

Section: Implementations Of Fair-rl For Non-societal Fairnessmentioning

confidence: 99%

Survey on Fair Reinforcement Learning: Theory and Practice

Gajane¹,

Saxena²,

Tavakol³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Studying effect of network parameters on its performance SVM [14] Fairness improvement Striving to balance user experience among users MDP [15] III. APPLICATIONS OF ML FOR NETWORK MANAGEMENT IN WMNS When maintaining a WMN, it is critical to pay attention to certain management-level issues that may compromise the security, integrity or the expected performance level of the system.…”

Section: Detecting and Alerting Users About Possible Attacksmentioning

confidence: 99%

“…However, the fundamental differences between wireless and wired mediums result in substandard performance of TCP over wireless networksespecially affecting TCP unfairness-as TCP favors flows with smaller number of hops in WMNs. To tackle this problem, authors of [15] suggested an approach where each TCP source models the state of the system as an MDP and uses Q-learning to learn the transition probabilities of the proposed MDP based on the observed variables. To maximize TCP fairness, each node hosting a TCP source takes actions according to the recommendations of the Q-learning algorithm and adjusts TCP parameters autonomously.…”

Section: Support Vector Machines (Svms)mentioning

confidence: 99%

An Overview of Machine Learning Approaches in Wireless Mesh Networks

Karunaratne

Gacanin

2019

IEEE Commun. Mag.

View full text Add to dashboard Cite

Wireless Mesh Networks (WMNs) have been extensively studied for nearly two decades as one of the most promising candidates expected to power the high bandwidth, high coverage wireless networks of the future. However, consumer demand for such networks has only recently caught up, rendering efforts at optimizing WMNs to support high capacities and offer high QoS, while being secure and fault tolerant, more important than ever. To this end, a recent trend has been the application of Machine Learning (ML) to solve various design and management tasks related to WMNs. In this work, we discuss key ML techniques and analyze how past efforts have applied them in WMNs, while noting some existing issues and suggesting potential solutions. We also provide directions on how ML could advance future research and examine recent developments in the field.

show abstract

“…Despite their influence on the real world, automated decision-making systems have come under scrutiny for potentially unfair outcomes between sub-groups separated based on protected data attributes, such as race, gender, and marital status [7,11,18,22,24,43]. The concept of fairness is applicable more broadly, including technical settings such as fair resource allocation in computer networks [3] and fair task assignment in crowdsourcing [6].…”

Section: Introductionmentioning

confidence: 99%