An Adaptive Authentication Based on Reinforcement Learning

Cui, Ziqi; Zhao, Yongqiang; Li, Chunxi; Zuo, Qiting; Zhang, Haipeng

doi:10.1109/icce-tw46550.2019.8991836

Cited by 6 publications

(6 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A four-factor authentication method proposal was also found, the fourth factor being the location-based factor [40]. The appearance of dynamic authentication methods [41][42][43] was an interesting find, as these methods adapt to distinct contexts.…”

Section: Discussionmentioning

confidence: 99%

Framework for the Comparison and Selection of Schemes for Multi-Factor Authentication

Velásquez

2021

CLEIej

View full text Add to dashboard Cite

Authentication is the process of verifying a user’s identity for them to access a system’s resources. An authentication factor is a piece of information used for this authentication. Three well-known groups of authentication factors exist: knowledge-based (what you know), possession-based (what you have) and inherence-based (what you are). Authentication schemes belonging to distinct authentication factors can be combined in a multi-factor manner to increase security. Although multiple multi-factor proposals are seen in literature, the absence of a method that allows a proper comparison and selection of these authentication methods, based on an application’s security requirements, can be observed. Existing frameworks for the analysis of authentication methods have been identified through the realization of a systematic literature review, but most of these focus on specific contexts and do not provide a generic enough solution. Due to the above, this research focuses on the creation of a recommendation framework that guides in the comparison and selection of single and multi-factor authentication schemes, considering both the application’s requirements and its context. This has been attained not only through the knowledge found in literature, but the experience from industry experts has been compiled as well through the collaboration with a multinational software development company. Consequently, the knowledge found in literature has been obtained from a systematic literature review, whereas the experience from industry experts was obtained through a survey and interviews. The framework proposal has been generated based on the above and has been validated through an expert panel and a case study methodology in collaboration with the partnered software development company. A tool prototype has been constructed as well. The result is a recommendation framework for the comparison and selection of authentication methods that can support this decision process in multiple contexts.

show abstract

Section: Discussionmentioning

confidence: 99%

Framework for the Comparison and Selection of Schemes for Multi-Factor Authentication

Velásquez

2021

CLEIej

View full text Add to dashboard Cite

show abstract

“…RL models interact with the environment to exercise rewards and penalties and automatically determine the ideal behavior to maximize the defined performance metric. This reward feedback in the RL model can be used to find the best action for classification [80][81][82], feature selection [38], or any other required decision in the system. Cui et al [80] used RL to propose an adaptive authentication scheme.…”

Section: Reinforcement Learningmentioning

confidence: 99%

“…This reward feedback in the RL model can be used to find the best action for classification [80][81][82], feature selection [38], or any other required decision in the system. Cui et al [80] used RL to propose an adaptive authentication scheme. They developed a multi-factor authentication method that uses different combinations of authentication models proportionate to www.videleaf.com the level of authentication confidence requirements.…”

Section: Reinforcement Learningmentioning

confidence: 99%

Using Machine Learning for Dynamic Authentication in Telehealth: A Tutorial

Hazratifard,

Gebali,

Mamun

2023

Prime Archives in Sensors

View full text Add to dashboard Cite

show abstract

“…RL models interact with the environment to exercise rewards and penalties and automatically determine the ideal behavior to maximize the defined performance metric. This reward feedback in the RL model can be used to find the best action for classification [ 80 , 81 , 82 ], feature selection [ 38 ], or any other required decision in the system. Cui et al [ 80 ] used RL to propose an adaptive authentication scheme.…”

Section: Machine Learning Models In Authentication Schemes Of Telehealthmentioning

confidence: 99%

“…This reward feedback in the RL model can be used to find the best action for classification [ 80 , 81 , 82 ], feature selection [ 38 ], or any other required decision in the system. Cui et al [ 80 ] used RL to propose an adaptive authentication scheme. They developed a multi-factor authentication method that uses different combinations of authentication models proportionate to the level of authentication confidence requirements.…”

Section: Machine Learning Models In Authentication Schemes Of Telehealthmentioning

confidence: 99%

Using Machine Learning for Dynamic Authentication in Telehealth: A Tutorial

Fard

Gebali

Mamun

2022

Sensors

View full text Add to dashboard Cite

Telehealth systems have evolved into more prevalent services that can serve people in remote locations and at their homes via smart devices and 5G systems. Protecting the privacy and security of users is crucial in such online systems. Although there are many protocols to provide security through strong authentication systems, sophisticated IoT attacks are becoming more prevalent. Using machine learning to handle biometric information or physical layer features is key to addressing authentication problems for human and IoT devices, respectively. This tutorial discusses machine learning applications to propose robust authentication protocols. Since machine learning methods are trained based on hidden concepts in biometric and physical layer data, these dynamic authentication models can be more reliable than traditional methods. The main advantage of these methods is that the behavioral traits of humans and devices are tough to counterfeit. Furthermore, machine learning facilitates continuous and context-aware authentication.

show abstract

An Adaptive Authentication Based on Reinforcement Learning

Cited by 6 publications

References 1 publication

Framework for the Comparison and Selection of Schemes for Multi-Factor Authentication

Framework for the Comparison and Selection of Schemes for Multi-Factor Authentication

Using Machine Learning for Dynamic Authentication in Telehealth: A Tutorial

Using Machine Learning for Dynamic Authentication in Telehealth: A Tutorial

Contact Info

Product

Resources

About