Collaborative Filtering with Graph Information: Consistency and Scalable Methods

By Nikhil S. Rao, Hsiang-Fu Yu, Pradeep Ravikumar, Inderjit S. Dhillon. 2015

302 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.

Citations in the corpus

Geometric Matrix Completion with Recurrent Multi-Graph Neural Networks (2017, methodology): “We closely followed the experimental setup of (Rao et al., 2015), using ﬁve standard datasets: Synthetic dataset from (Kalofolias et al., 2014), MovieLens (Miller et al., 2003), Flixster (Jamali & Ester, 2010), Douban (Ma et al., 2011), and YahooMusic (Dror et al., 2012).”
Geometric Deep Learning: Going beyond Euclidean data (2016, background): “structure into matrix completion problems [94], [95], [96], [97] in the form of column- and row graphs representing similarity of users and items, respectively (see Figure 4).”
Propagation Model Search for Graph Neural Networks (2020, methodology): “Take recommender systems (Monti, Bronstein, and Bresson 2017; Rao et al. 2015) as an example, data can be represented as a 3 - way tensor H ∈ R N 1 × N 2 × d .”
Disentangled Graph Collaborative Filtering (2020, methodology): “Prior efforts like HOP-Rec [45] and GRMF [27] apply idea of unsupervised representation learning — that is, connected nodes have similar representations — to smooth user and item representations.”
LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation (2020, unknown): “• GRMF [30].”
Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction (2016, methodology): “As an application of Theorem 1 from [15] and Corollary 1, when R f ( F ) = (cid:107) F 2 F , we can relate T AR to a weighted nuclear norm: where B = US 1 / 2 and Σ − 1 AR = USU (cid:62) is the eigen-decomposition of Σ − 1 AR .”
UvA-DARE ( Digital Academic Repository ) Graph Convolutional Matrix Completion (2017, unknown): “[25], we map the additional information onto feature vectors for users and movies, and compare the performance of our model with (GC-MC+Feat) and without the inclusion of these features (GC-MC) .”