Nonparametric Latent Feature Models for Link Prediction

By Kurt T. Miller, T. Griffiths, Michael I. Jordan. 2009

436 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

A semantic matching energy function for learning with multi-relational data (2013, background): “Let C denote the dictionary which includes all entities and relation types, and let R ⊂ C be the subset of entities which are relation types.”
Exploiting compositionality to explore a large space of model structures (2012, methodology): “We note that our initialization procedure generalizes “tricks of the trade” whereby complex models are initialized from simpler ones (Kemp et al., 2006; Miller et al., 2009).”
Cross View Link Prediction by Learning Noise-resilient Representation Consensus (2017, background): “Besides, various latent variable models [3] [29] [17] have been proposed to model the relationship between nodes.”
A latent factor model for highly multi-relational data (2012, methodology): “To share parameters between relations, [9, 24, 14, 28] and [10] build models that cluster not only entities but relations as well.”
Embedding Networks with Edge Attributes (2018, methodology): “Some other methods use Bayesian models to learn inherent structure of the network [20].”
Dropout Training of Matrix Factorization and Autoencoder for Link Prediction in Sparse Graphs (2015, methodology): “Bayesian models, such as MMSB [2, 15, 26] are another family of latent factor models for studying network structures.”
Infinite multiple membership relational modeling for complex networks (2011, unknown): “As in existing multiple membership models [11, 12] we will as sume an unbounded number of latent features.”
Probabilistic machine learning and artificial intelligence (2015, background): “…to some subset of a large number of potential communities (for example, as defined by different families, workplaces, schools, hobbies, and so on) rather than a single community, and the probability of friendship between two people depends on the number of overlapping communities they have 42 .”
Translating Embeddings for Modeling Multi-relational Data (2013, background): “…natural extensions of these approaches to the multi-relational domain such as non-parametric Bayesian extensions of the stochastic blockmodel [7, 10, 17] and models based on tensor factorization [5] or collective matrix factorization [13, 11, 12], many of the most recent approaches have focused…”
Infinite Latent SVM for Classification and Multi-task Learning (2011, methodology): “1 Indian Buffet Process Indian buffet process (IBP) was proposed in [12] and has been successfully applied in various fields, such as link prediction [22] and multi-task learning [23].”
Semi-supervised Penalized Output Kernel Regression for Link Prediction (2011, unknown): “The main approaches devoted to this task fall into two families: probabilistic graphical models (Miller et al., 2009; Taskar et al., 2003) provide posterior probabilities of links while kernel or similarity based methods take beneﬁt from the versatility of kernels to encode various structured…”
LRBM: A Restricted Boltzmann Machine Based Approach for Representation Learning on Linked Data (2014, background): “W is usually constrained to be symmetric for modeling symmetric (i.e. undirected) linked data and be asymmetric for modeling directed linked data.”
Deep Generative Models for Relational Data with Side Information (2017, methodology): “Extensions of the latent space model include stochastic blockmodels (Nowicki & Snijders, 2001), and variants thereof (Miller et al., 2009; Airoldi et al., 2008; Latouche et al., 2011), which can learn node embeddings that are interpretable (e.”
Explainable Product Search with a Dynamic Relation Embedding Model (2019, methodology): “on matrix and the item-item co-occurrence matrix with shared item embeddings. There is also another line of studies trying to tackle the problem with non-parametric Bayesian frameworks. Miller et al. [41] experiment with a non-parametric Bayesian model for link prediction ACM Transactions on Informa”
Weisfeiler-Lehman Neural Machine for Link Prediction (2017, methodology): “Popular latent feature-based methods include: matrix factorization [23]; ranking methods [24] which treat link prediction as a learning to rank problem; and stochastic block models [5, 25], which assume that nodes have latent groups and links are determined by the group memberships of nodes.”