Rank Machine Learning
noun
Definition 1: The rank of a matrix; the dimension of the subspace spanned by its rows or columns [Bakanach 2023/2024].
Example in context: “This means that the rank of the matrix Y, say r0, is lower than the number of samples.” [Ribaud et al. 2024]
Definition 2: The position of a document or object in an ordering by decreasing relevance or score; in evaluation, ranks are used to compute ranking metrics such as NDCG [Machine Learning Glossary].
Example in context: “We expect the optimal context size to depend on the average rank that ground-truth documents tend to receive, and for models of similar ranking effectiveness, this would primarily be determined by the characteristics of the dataset.” [Zerveas et al. 2023]
Synonym: position (sense 2)
Related terms: matrix rank, low-rank approximation, ranking, reranking, relevance score, NDCG, top-k