Multitask learning meets tensor factorization: task imputation via convex optimization

By Kishan Wimalawarne, Masashi Sugiyama, Ryota Tomioka. 2014

83 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 Survey on Multi-Task Learning (2017, methodology): “Multilinear multi-task methods proposed in [115], [116] can handle this situation by”
Deep Multi-task Representation Learning: A Tensor Factorisation Approach (2016, methodology): “Our proposed framework applies to this multiclass MDL setting where existing single-output MTL methods [Evgeniou and Pontil, 2004, Argyriou et al., 2008, Kumar and Daum´e III, 2012, Romera-paredes et al., 2013, Wimalawarne et al., 2014] do not.”
A Unified Perspective on Multi-Domain and Multi-Task Learning (2014, methodology): “Then they follow the line of Argyriou et al. (2008) to impose a variety of regularisations on the mentioned tensor, such as sum of the ranks of the matriciations of the tensors (Romera-paredes et al., 2013) and scaled latent trace norm (Kishan Wimalawarne & Tomioka, 2014).”
Tensor Based Knowledge Transfer Across Skill Categories for Robot Control (2017, methodology): “A few MTL studies used low-rank tensors for knowledge sharing across tasks with a structured description, rather than a simple index [Wimalawarne et al., 2014].”
Tensor Networks for Dimensionality Reduction and Large-scale Optimization: Part 2 Applications and Future Perspectives (2017, background): “One of the important and useful tensor norms is the tensor nuclear norm [Liu et al., 2013] or the (overlapped) trace norm [Wimalawarne et al., 2014], which can be defined for a tensor W P RI1ˆ¨¨¨ˆIN as }W}˚ = N ÿ n=1 }W(n)}˚, (2.15) where }W(n)}˚ = ř i σi, σi denotes the ith singular value of W(n).”
Canonical Tensor Decomposition for Knowledge Base Completion (2018, background): “…along each of its modes to obtain matricizations, and either regularize by the sum of trace norms of the matricizations (Tomioka et al., 2010) or write the original tensor as a sum of tensors T k , regularizing their respective k th matricizations with the trace norm (Wimalawarne et al., 2014).”
Trace Norm Regularised Deep Multi-Task Learning (2016, background): “For example, sum of the trace norms on all matriciations [Romera-paredes et al., 2013], and scaled latent trace norm [Wimalawarne et al., 2014].”