English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Missing data are an unavoidable complication frequently encountered in manycausal discovery tasks. When a missing process depends on the missing valuesthemselves (known as self-masking missingness), the recovery of the jointdistribution becomes unattainable, and detecting the presence of suchself-masking missingness remains a perplexing challenge. Consequently, due tothe inability to reconstruct the original distribution and to discern theunderlying missingness mechanism, simply applying existing causal discoverymethods would lead to wrong conclusions. In this work, we found that the recentadvances additive noise model has the potential for learning causal structureunder the existence of the self-masking missingness. With this observation, weaim to investigate the identification problem of learning causal structure frommissing data under an additive noise model with different missingnessmechanisms, where the `no self-masking missingness' assumption can beeliminated appropriately. Specifically, we first elegantly extend the scope ofidentifiability of causal skeleton to the case with weak self-maskingmissingness (i.e., no other variable could be the cause of self-maskingindicators except itself). We further provide the sufficient and necessaryidentification conditions of the causal direction under additive noise modeland show that the causal structure can be identified up to an IN-equivalentpattern. We finally propose a practical algorithm based on the abovetheoretical results on learning the causal skeleton and causal direction.Extensive experiments on synthetic and real data demonstrate the efficiency andeffectiveness of the proposed algorithms.

Identification of Causal Structure in the Presence of Missing Data with Additive Noise Model