Learning to rank: from pairwise approach to listwise approach

By Zhe Cao, Tao Qin, Tie-Yan Liu, Ming-Feng Tsai, Hang Li. 2007

2,421 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Information Retrieval & Search, Signal Processing

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Information Systems

In Information Retrieval & Search

In Signal Processing

Citations in the corpus

VD-BERT: A Unified Vision and Dialog Transformer with BERT (2020, methodology): “We also show three other apparent attention patterns: attentions that a token puts to its previous token, to itself, and to the next token.”
Mean Field Multi-Agent Reinforcement Learning (2018, neutral): “Similar approaches can be found in factorization machine (Rendle, 2012) and learning to rank (Cao et al., 2007).”
End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension (2016, background): “The softmax in the ﬁnal layer serves as the list-wise ranking module similar in spirit to (Cao et al. 2007).”
Predicting a Correct Program in Programming by Example (2015, methodology): “This problem formulation is similar to the problem formulation of listwise approaches for learning-to-rank [3, 27].”
Optimal full ranking from pairwise comparisons (2021, background): “This highdimensional statistical estimation problem has important applications in many areas such as recommendation systems [1, 7], sports and gaming [2, 17, 23, 40, 41, 49], web search [16, 19], social choices [33, 36, 38, 39, 47], psychology [13, 34, 52], information retrieval [8, 32], etc.”
Tapping into knowledge base for concept feedback: leveraging conceptnet to improve search results for difficult queries (2012, background): “, learning-to-rank, such as ListNet [8]) as future work.”
Entity Linking with a Knowledge Base: Issues, Techniques, and Solutions (2015, methodology): “entities: the pairwise framework Ranking Perceptron [117] and the listwise framework ListNet [118].”
Ranking Policy Gradient (2019, methodology): “Theorem 3 ((Cao et al., 2007), Theorem 6) Given the action values q = [ Q 1 , ..., Q m ] , the probability of action i to be chosen (i.e. to be ranked on the top of the list) is: where φ ( ∗ ) is any increasing, strictly positive function.”
Direct Learning to Rank and Rerank (2018, background): “…2013), ranking water pipes in order of vulnerability (as considered by Li et al, 2013), other areas of information retrieval (Xu, 2007; Cao et al, 2007; Matveeva et al, 2006; Lafferty and Zhai, 2001; Li et al, 2007) and in almost any domain where one measures the quality of results by…”
Learning to Rank Short Text Pairs with Convolutional Deep Neural Networks (2015, methodology): “The third method, referred to as a listwise approach [6], treats a query with its list of candidates as a single instance in learning, thus able to capture considerably more information about the ground truth ordering of input candidates.”
List-wise learning to rank with matrix factorization for collaborative filtering (2010, methodology): “2 Top One Probability As proposed in [2], the top one probability for an item rated Rij in user i’s ranking list li (e.”
Modelling input texts: from Tree Kernels to Deep Learning (2015, methodology): “The third method, referred to as a listwise approach [24], treats a query with its list of candidates as a single instance in learning, thus able to capture considerably more information about the ground truth ordering of input candidates.”
Relative attributes (2011, methodology): “While alternative list-based learning to rank techniques are available [23], we choose the pairwise learning technique as described in Section 3.”