Transfer bounds for linear feature learning

By Andreas Maurer. 2009

102 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

Continual learning: a feature extraction formalization, an efficient algorithm, and fundamental obstructions (2022, background): “Some recent theoretical works [18, 19, 26, 36, 20, 35, 9] provide generalization and sample complexity guarantees for certain formalizations of multi-task learning based on the existence of a good representation.”
A PAC-Bayesian bound for Lifelong Learning (2013, methodology): “In a case of linear regression, Maurer (2009) used this assumption to prove a generalization bound in the PAC framework, by us-ing the concept of environment of tasks from Baxter (2000) and Rademacher complexity.”
Sparse coding for multitask and transfer learning (2012, methodology): “The precursors of the analysis presented here are (Maurer & Pontil, 2010) and (Maurer, 2009).”
Multi-task and Lifelong Learning of Kernels (2015, background): “Along that line, Maurer in [19] proved that learning a common low-dimensional representation in the case of lifelong learning of linear least-squares regression tasks is beneficial.”
An Information-Theoretic Approach to Transferability in Task Transfer Learning (2019, background): “There have been theoretical studies that focus on task relatedness ([8, 9, 10, 11]).”
Lifelong Learning with Weighted Majority Votes (2016, background): “These settings were theoretically analyzed in a series of works [18, 19, 23, 21] that have demonstrated that information transfer can lead to provable sample complexity reductions compared to solving each task independently.”
Learning To Learn Around A Common Mean (2018, methodology): “we do not work on what in literature is called future empirical risk [12, 20]), but we sacrifice a subset of them for testing.”
Multilinear Multitask Learning (2013, background): “Copyright 2013 by the author(s). learning process, which leads to better generalization if the tasks are related (Ando and Zhang, 2005; Baxter, 2000; Maurer, 2009; Maurer and Pontil, 2012; Maurer et. al, 2013).”
Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theory (2017, background): “Other recent theoretical approaches to meta or multitask learning (e.g. Maurer (2005; 2009); Ruvolo & Eaton (2013); Maurer et al. (2016); Alquier et al. (2017)) provide increasingly general bounds but have not led to practical algorithms for neural networks.”
A Vector-Contraction Inequality for Rademacher Complexities (2016, background): “Examples are given in [12] or [14], but see also the next section.”
How to Train Your MAML to Excel in Few-Shot Classification (2021, background): “A “meta-model” summarizes the common characteristic of tasks and generalizes them to those novel but related tasks [19, 58, 59].”
Incremental Learning-to-Learn with Statistical Guarantees (2018, methodology): “This quantity plays a key role in identifying the environments ρ for which it is favorable to adopt a LTL strategy instead of solving the tasks independently, see [20] for a discussion.”
Few-shot learning with adaptively initialized task optimizer: a practical meta-learning approach (2019, background): “A meta-model characterizes the task commonality and generalizes its prediction to those unseen tasks from a related environment (Maurer 2009; Maurer et al. 2016; Denevi et al. 2018).”
Fairness in Machine Learning (2020, methodology): “Notice that the ﬁrst bound in Theorem 2 improves Theorem 2 in [125].”
Multitask Sparsity via Maximum Entropy Discrimination (2011, methodology): “Since generalization guarantees in multitask settings have already been provided for other algorithms (Ando and Zhang, 2005; Maurer, 2006, 2009), this may be a fruitful line of work.”
Learning-to-Learn Stochastic Gradient Descent with Biased Regularization (2019, background): “, 2018a;b; Pentina & Urner, 2016) has received limited attention and is less developed than standard LTL approaches, in which the data are processed in one batch as opposed to incrementally, see for instance (Baxter, 2000; Maurer, 2009; Maurer et al., 2013; 2016; Pentina & Lampert, 2014).”