Convex multi-task feature learning

By Andreas Argyriou, Theodoros Evgeniou, Massimiliano Pontil. 2008

1,616 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Mechanics, Computer Networks and Communications, Representation & Self-Supervised Learning, Optimization & Training

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

In Computer Networks and Communications

In Representation & Self-Supervised Learning

In Optimization & Training

Citations in the corpus

Cross-Stitch Networks for Multi-task Learning (2016, methodology): “The term multi-task learning (MTL) itself has been broadly used [2, 14, 28, 42, 54, 55] as an umbrella term to include representation learning and selection [4, 13, 31, 37], transfer learning [39, 41, 56] etc.”
Semi-Cyclic Stochastic Gradient Descent (2019, background): “…features or low-rank structure (Ando & Zhang, 2005), shared deep features (e.g. Caruana, 1997), shared kernel or low nuclearnorm structure (e.g. Argyriou et al., 2008; Amit et al., 2007), low-norm additive structure (e.g. Evgeniou & Pontil, 2004), or graph-based relatedness (e.g. Evgeniou et…”
LEAF: A Benchmark for Federated Settings (2018, background): “This paradigm is also amenable to federated settings [30] but, contrary to realistic federated networks, is usually explored in regimes with small numbers of tasks and samples, e.g., the popular Landmine Detection [38, 26, 36, 30], Computer Survey [2, 1, 14] and Inner London Education Authority Scho”
A Survey on Multi-Task Learning (2017, methodology): “Different from multi-layer feedforward neural networks which are connectionist models, the multi-task feature learning (MTFL) method [11], [12] is formulated under the regularization framework with the objective function as”
Federated Multi-Task Learning (2017, methodology): “Before proposing our federated method for solving (1), we make the following observations: • Observation 1 : In general, (1) is not jointly convex in W and Ω , and even in the cases where (1) is convex, solving for W and Ω simultaneously can be difﬁcult [5].”
Kernels for Vector-Valued Functions: a Review (2011, methodology): “The authors of [4] discuss how the above model is equivalent to considering a kernel of the form”
Selective Experience Replay for Lifelong Learning (2018, background): “While episodic memory could involve generative processes (Goodfellow et al. 2014) or compression (Argyriou, Evgeniou, and Pontil 2008), in this work we focus speciﬁcally on the case of selecting complete experiences for storage.”
A PAC-Bayesian bound for Lifelong Learning (2013, methodology): “(Argyriou et al., 2008; Kumar & Daumé III, 2012), which have also successfully been applied in the lifelong learning setting (Ruvolo & Eaton, 2013).”
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).”
Successor Features for Transfer in Reinforcement Learning (2016, background): “(Generalized Policy Improvement) Let π1, π2, ..., πn be n decision policies and let Q̃π1 , Q̃π2 , ..., Q̃πn be approximations of their respective action-value functions such that |Qπi(s, a)− Q̃πi(s, a)| ≤ for all s ∈ S, a ∈ A, and i ∈ {1, 2, ..., n}.”
An embarrassingly simple approach to zero-shot learning (2015, background): “For example if L is the sum of hinge losses, and Ω is the Frobenius norm, this would lead to a standard support vector machine (SVM), but one can consider other loss functions such as logistic loss, and other regularizers, such as the trace norm, leading to multitask learning methods [2, 32].”
A Convex Formulation for Learning Task Relationships in Multi-Task Learning (2010, methodology): “For our performance measure, we use the measure of percentage explained variance from [1], which is deﬁned as the percentage of one minus nMSE.”
Domain Generalization for Object Recognition with Multi-task Autoencoders (2015, positive): “Our strategy is motivated by prior work demonstrating that learning from multiple related tasks can improve performance on a novel, yet related, task – relative to methods trained on a single-task [1, 3, 8, 34].”
Joint covariate selection and joint subspace selection for multiple classification problems (2010, background): “The general problem of jointly estimating models from multiple, related data sets is often referred to as “transfer learning” or “multi-task learning” in the machine learning literature (Maurer 2006; Ben-David and Schuller-Borbely 2008; Argyriou et al. 2008; Jebara 2004; Evgeniou and Pontil 2004; To”
Curriculum learning of multiple tasks (2014, methodology): “proposed a method to learn such representations using sparsity regularization in [1].”