The Kendall Rank Correlation Coefficient

564 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Multinomial Distribution Learning for Effective Neural Architecture Search (2019, unknown): “To get some intuitive sense of the hypothesis, we introduce the Kendall rank correlation coefﬁcient, a.k.a. Kendall’s τ [1].”
Design Space for Graph Neural Networks (2020, methodology): “Finally, we use Kendall rank correlation [1] to compute the similarity score between the rankings of the M anchor models, which is reported as task similarity.”
The EMPATHIC Framework for Task Learning from Implicit Human Feedback (2020, methodology): “To measure performance, Kendall’s rank correlation coefficient τ ∈ [−1, 1] [41] is used to compare a human proxy’s inferred ranking with ground truth (a higher τ value indicates a higher correlation between two rankings).”