Corpus ID: 9293483

Multi-relational Learning Using Weighted Tensor Decomposition with Modular Loss

@article{London2013MultirelationalLU,
  title={Multi-relational Learning Using Weighted Tensor Decomposition with Modular Loss},
  author={Ben London and Theodoros Rekatsinas and Bert Huang and L. Getoor},
  journal={ArXiv},
  year={2013},
  volume={abs/1303.1733}
}
We propose a modular framework for multi-relational learning via tensor decomposition. In our learning setting, the training data contains multiple types of relationships among a set of objects, which we represent by a sparse three-mode tensor. The goal is to predict the values of the missing entries. To do so, we model each relationship as a function of a linear combination of latent factors. We learn this latent representation by computing a low-rank tensor decomposition, using quasi-Newton… Expand
Tensor factorization for relational learning
Iterative Splits of Quadratic Bounds for Scalable Binary Tensor Factorization
Multi-tensor Completion with Common Structures
Large-scale factorization of type-constrained multi-relational data
Zero-Truncated Poisson Tensor Factorization for Massive Binary Tensors
Logistic Tensor Factorization for Multi-Relational Data
Multi-Task Metric Learning on Network Data
Using Joint Tensor Decomposition on RDF Graphs
Probabilistic Latent-Factor Database Models
...
1
2
...

References

SHOWING 1-10 OF 25 REFERENCES
A Three-Way Model for Collective Learning on Multi-Relational Data
Link Pattern Prediction with tensor decomposition in multi-relational networks
Fast maximum margin matrix factorization for collaborative prediction
Multilinear algebra for analyzing data with multiple linkages
Collaborative Filtering on a Budget
...
1
2
3
...