Identification of Outliers in Graph Signals

Gopalakrishnan, Karthik; Li, Max Z.; Balakrishnan, Hamsa

doi:10.1109/cdc40024.2019.9029478

Cited by 10 publications

(9 citation statements)

References 14 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Function N → F (2, N ) is nondecreasing, but it is not strictly concave, because as demonstrated by Eq. (20) its neighboring values can be equal. The proof of these two hypotheses may be very difficult, but they are not essential for (27), which represent the exact and approximated values of F (n, N ), respectively: a) the absolute error, b) the relative error, and c) the dependance of certain relative errors on n and N .…”

Section: ) Asymptoticsmentioning

confidence: 99%

See 1 more Smart Citation

TVOR: Finding Discrete Total Variation Outliers Among Histograms

Banić¹,

Elezović

2021

IEEE Access

View full text Add to dashboard Cite

In this paper, we respond to a critique of one of our papers previously published in this journal, entitled "TVOR: Finding Discrete Total Variation Outliers among Histograms". Our paper proposes a method for smoothness outliers detection among histograms by using the relation between their discrete total variations (DTV) and their respective sample sizes. In this response, we demonstrate point by point that, contrary to its claims, the critique has not found any mistakes or problems in our paper, either in the used datasets, methodology, or in the obtained top outlier candidates. On the contrary, the critique's claims can easily be shown to be mathematically unfounded, to directly contradict the common statistical theorems, and to go against well established demographic terms. Exactly this is done in the reply here by providing both theoretical and experimental evidence. Additionally, due to critique's compalint, a more extensive research on top outlier candidate, i.e. the Jasenovac list is conducted and in order to clear any of the critique's doubts, new evidence of its problematic nature unseen in other lists are presented. This reply is accompanied by additional theoretical explanations, simulations, and experimental results that not only confirm the earlier findings, but also present new data. The source code is at https://github.com/DiscreteTotalVariation/TVOR.

show abstract

Section: ) Asymptoticsmentioning

confidence: 99%

“…Single-value properties similar to total variation in terms of simplicity, such as skewness, have already been used for outlier detection [18]. As a matter of fact, total variation has also found application in tasks such as classification [19] and outlier detection for graph signals [20].…”

Section: Introductionmentioning

confidence: 99%

TVOR: Finding Discrete Total Variation Outliers Among Histograms

Banić¹,

Elezović

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The control of delays in airport networks has been a longstanding problem [16]. The use of low-dimensional, human-interpretable performance measures (e.g., total delay, spatial distribution of delays, see [17]) in a large-scale system makes this an ideal setting for our methodology.…”

Section: Prior Workmentioning

confidence: 99%

“…The choice of x 1 is motivated by the fact that in positive signal-generating systems, this metric captures the total magnitude of signals across the entire system (e.g., the total delay in an airport network). The choice of TV(x) = x Lx reflects the fact that TV can be interpreted as a measure of signal smoothness, and has been used for outlier detection in graph signals [17]. Our projection-based control framework is agnostic to the specific choice of metrics; any reasonable set of low-dimensional metrics that captures key system performance characteristics may be used.…”

Section: B Projection-based Controlmentioning

confidence: 99%

Approximate Projection-Based Control of Networks

Gopalakrishnan

Balakrishnan

2020

2020 59th IEEE Conference on Decision and Control (CDC)

Self Cite

View full text Add to dashboard Cite

Modern infrastructures such as transportation and communication networks are large-scale systems with complex dependence structures between various sub-systems. Human-interpretable performance targets in such systems are often represented in terms of lower-dimensional projections of the high-dimensional state space. We consider the problem of designing control strategies for high-dimensional systems that lack a detailed model. To do so, we leverage the ability of copulas to represent dependant structures in high-dimensional data, and approximate the state space model through inverse sampling. We demonstrate the applicability of the control policies obtained from our methodology through a data-driven case study of controlling flight delays within the US air transportation network.

show abstract

“…One unanswered question is: what can we gain if we analyze the air transportation complex system by combining information about both structure and dynamics? A series of recent studies on outlier detection have demonstrated the advantage of investigating airport networks taking into account both network structure and dynamics [10,11]. e authors constructed the airport networks based on the correlations between airport delays.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Study on Flight Delay Networks of the USA and China

Wang

Hong-feng

et al. 2020

Journal of Advanced Transportation

View full text Add to dashboard Cite

Recent studies have characterized the structures of air transport network in different countries and regions using complex network metrics. These studies coincided with the trend of increasingly available large empirical flight datasets that enable researchers to investigate the dynamics of the system, such as the propagation of flight delay. However, linking network structure with network dynamics remains a challenging task. In this paper, we proposed a method to construct flight delay networks from operational data. We provided a detailed comparison of the key structural properties of the flight delay networks in the United States and China. The comparisons of betweenness centrality of delay networks and flight networks show the advantage of the proposed method. We further found that airports in similar geographical locations do exhibit similar delay patterns in both countries. To explore the underlying mechanisms, the Multifractal Detrended Fluctuation Analysis (MF-DFA) is applied to the flights’ delay time series at both the airport level and network level. Singularity spectra analyses reveal the fundamental characteristics of the airport systems and air transportation system. Our findings contribute to the understanding of structure and dynamics of air transportation systems.

show abstract

Identification of Outliers in Graph Signals

Cited by 10 publications

References 14 publications

TVOR: Finding Discrete Total Variation Outliers Among Histograms

TVOR: Finding Discrete Total Variation Outliers Among Histograms

Approximate Projection-Based Control of Networks

A Comparative Study on Flight Delay Networks of the USA and China

Contact Info

Product

Resources

About