Asymptotic Performance In Heterogeneous Human-machine Inference Networks

Chen, Quan; Geng, Baocheng; Varshney, Pramod K.

doi:10.1109/ieeeconf51394.2020.9443353

Cited by 8 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, M . To emulate the actions humans take to incorporate the data gathered from side information and observations into their decision-making process, two operations are employed in this work which are OR operation and AND operation [3].…”

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

“…where w t m is the side information indicating whether H 1 is present or not and is assumed to be binary for human 3 The side information refers to the additional information owned by human sensors which could come from previous professional experience or other sources. sensor m. The accuracy of side information is denoted as P r(w t m = 1|H 1 ) = β m,side , and P r(w t m = 1|H 0 ) = γ m,side .…”

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

“…We assume that the side information {w t m } M m=1 is independent among different human sensors. Given the side information, the likelihoods of e t m given H 1 and H 0 are shown, respectively, as [3] f…”

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

“…In high stake scenarios where human lives and assets are at risk, automatic physical sensor-only decisionmaking may not be sufficient [1]- [3]. Further, in some circumstances, such as remote sensing and emergency access systems, humans may possess additional side information in addition to the common observations available from both physical sensors and humans.…”

Section: Introductionmentioning

confidence: 99%

“…For instance, the authors of [4], [5] studied decision fusion performance when the individual human agents use different thresholds modeled as random variables to make local decisions regarding a given phenomenon of interest (PoI). The authors in [3] proposed a hybrid system that consists of multiple human sub-populations, with the thresholds of each sub-population characterized by non-identically distributed random variables and a limited number of machines (physical sensors) whose exact values of thresholds are known. For such a hybrid system, they derived the asymptotic performance at the fusion center in terms of Chernoff information.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Human-machine Hierarchical Networks for Decision Making under Byzantine Attacks

Chen¹,

Geng²,

Han³

et al. 2023

Preprint

View full text Add to dashboard Cite

This paper proposes a belief-updating scheme in a human-machine collaborative decision-making network to combat Byzantine attacks. A hierarchical framework is used to realize the network where local decisions from physical sensors act as reference decisions to improve the quality of human sensor decisions. During the decisionmaking process, the belief that each physical sensor is malicious is updated. The case when humans have side information available is investigated, and its impact is analyzed. Simulation results substantiate that the proposed scheme can significantly improve the quality of human sensor decisions, even when most physical sensors are malicious. Moreover, the performance of the proposed method does not necessarily depend on the knowledge of the actual fraction of malicious physical sensors. Consequently, the proposed scheme can effectively defend against Byzantine attacks and improve the quality of human sensors' decisions so that the performance of the human-machine collaborative system is enhanced.

show abstract

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

Section: B Human Sensors With Side Informationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%