An Algorithm for Transfer Learning in a Heterogeneous Environment

By Andreas Argyriou, Andreas Maurer, Massimiliano Pontil. 2008

86 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mechanics

How researchers see this paper

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

In Artificial Intelligence

In Computational Mechanics

Citations in the corpus

Kernels for Vector-Valued Functions: a Review (2011, methodology): “A different approach is taken in [4] and [5].”
Learning Task Grouping and Overlap in Multi-task Learning (2012, methodology): “A starting point for our method is the assumption that task parameters lie in a low dimensional sub-space (Argyriou et al., 2008a).”
Sparse coding for multitask and transfer learning (2012, positive): “Our work combines ideas from sparse coding (Olshausen & Field, 1996), multitask learning (Ando & Zhang, 2005; Argyriou, Evgeniou, Pontil, 2008; Argyriou, Maurer, Pontil, 2008; Ben-David & Schuller, 2003; Caruana, 1997; Evgeniou, Micchelli, Pontil, 2005; Maurer, 2009) and learning to learn (Baxter,…”
Learning with Whom to Share in Multi-task Feature Learning (2011, methodology): “In particular, we contrast our work to a recent technique proposed in (Argyriou et al., 2008b), where the problem of inferring task clusters is cast as learning a set of kernel functions, one for each group.”
Multi-output learning via spectral filtering (2012, methodology): “The results of Argyriou et al. (2008a) were achieved using 75% of the data for training and adopting a 15-fold cross-validation to select the regularizing parameter.”
On Spectral Learning (2010, background): “trace norm is the `1 norm on the singular values and hence there is an analogy to regularization of a vector variable with the `1 norm, which is often used to obtain sparse solutions, see Candès and Recht (2008) and reference therein.”
Multilinear Multitask Learning (2013, methodology): “(5) When N = 2 problem (5) is equivalent to the one proposed in (Argyriou et al., 2008a).”
OTL: A Framework of Online Transfer Learning (2010, background): “Moreover, according to different feature representation, TL can be classiﬁed as homogeneous vs. heterogeneous TL (Argyriou et al., 2008) where the feature spaces of source and target domains can be diﬀerent.”
Feature-based transfer learning with real-world applications (2010, background): “As described above, most research works on model task correlation are from the context of multi-task learning [18, 11, 7, 83, 28, 224].”
Active Multi-task Learning via Bandits (2015, methodology): “[4] and Jacob et al.”
Local Rademacher Complexity-based Learning Guarantees for Multi-Task Learning (2016, background): “Transfer learning (Pan and Yang, 2010) and, in particular, Multi-Task Learning (MTL) (Caruana, 1997) leverage such underlying common links among a group of tasks, while respecting the tasks’ individual idiosyncrasies to the extent warranted. This is achieved by phrasing the learning process as a joi”
Convex Learning of Multiple Tasks and their Structure (2015, methodology): “This idea has motivated a variety of methods, including frequentist [25, 3, 4] and Bayesian methods (see e.g. [1] and references therein), with connections to structured learning [6, 34].”
Vector Field Learning via Spectral Filtering (2010, background): “Starting from the work of [7], related ideas have been developed in the context of regularization methods [8], Gaussian processes [9,10].”
A Survey on Transfer Learning (2010, background): “In a follow-up work, Argyriou et al. [41] proposed a spectral regularization framework on matrices for multitask structure learning.”
Flexible Modeling of Latent Task Structures in Multitask Learning (2012, methodology): “Apart from the prior work on multitask learning discussed in Section 1, our model is based on a somewhat similar motivation as the model proposed in (Argyriou et al., 2008).”