Learning to learn with the informative vector machine

By Neil D. Lawrence, John C. Platt. 2004

393 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Control and Systems Engineering, Representation & Self-Supervised Learning

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 Control and Systems Engineering

In Representation & Self-Supervised Learning

Citations in the corpus

An Overview of Multi-Task Learning in Deep Neural Networks (2017, background): “[Lawrence and Platt, 2004] extend Gaussian processes (GP) to MTL by inferring parameters for a shared covariance matrix.”
Towards a Neural Statistician (2016, background): “For example since the instance encoder can approximate a polynomial on a compact domain, and so can the post-pooling network, a statistic network can approximate any moment of a distribution.”
Recasting Gradient-Based Meta-Learning as Hierarchical Bayes (2018, background): “Hierarchical Bayesian models have a long history of use in both transfer learning and domain adaptation ( e.g., Lawrence & Platt, 2004; Yu et al., 2005; Gao et al., 2008; Daumé III, 2009; Wan et al., 2012).”
Probabilistic Model-Agnostic Meta-Learning (2018, background): “Hierarchical Bayesian models are a long-standing approach for few-shot learning that naturally allow for the ability to reason about uncertainty over functions [39, 7, 25, 43, 12, 4, 41].”
Bayesian Model-Agnostic Meta-Learning (2018, background): “There have been many studies in the past that formulate meta-learning and learning-to-learn from a probabilistic modeling perspective (Tenenbaum, 1999; Fe-Fei et al., 2003; Lawrence & Platt, 2004; Daumé III, 2009).”
An embarrassingly simple approach to zero-shot learning (2015, background): “machine learning problems, and has received several names, such as learning to learn [23] or inductive transfer [7, 31, 33].”
Causal Transfer in Machine Learning (2015, background): “For instance, Lawrence and Platt [2004] introduce a model for MTL in which the mapping f k in each task k ∈ { 1 , . . . , D, T } is drawn independently from a common Gaussian Process (GP), and the likelihood of the latent functions depends on a shared parameter θ .”
SMIL: Multimodal Learning with Severely Missing Modality (2021, methodology): “Previous methods studied meta-learning from the perspective of metric learning (Koch 2015; Vinyals et al. 2016; Sung et al. 2018; Snell, Swersky, and Zemel 2017) or probabilistic modeling (Fe-Fei et al. 2003; Lawrence and Platt 2004).”
Multi-task Gaussian Process Prediction (2007, background): “Within the GP literature, [14, 15, 16, 17, 18] give models where the covariance matrix of the full (noiseless) system is block diagonal, and each of the M blocks is induced from the same kernel function.”
A Survey on Transfer Learning (2010, methodology): “approach [45], [46], [47], [48], [49], which assumes that the”
Large-Scale Personalized Human Activity Recognition Using Online Multitask Learning (2013, background): “Bao and Intille [1] and Pärkkä et al. [2] used decision trees for classification.”
Multi-task Reinforcement Learning in Partially Observable Stochastic Environments (2009, background): “Most of the work follows a hierarchical Bayesian approach, which assumes that the parameters (models) for each task are sampled from a common prior distribution, such as a Gaussian distribution specified by unknown hyperparameters (Lawrence and Platt 2004; Yu et al. 2003).”
Invariant Models for Causal Transfer Learning (2015, background): “For instance, Lawrence and Platt (2004) introduce a model for MTL in which the mapping f k in each task k ∈ { 1 , . . . , D, T } is drawn independently from a common Gaussian Process (GP), and the likelihood of the latent functions depends on a shared parameter θ .”
Convex Discriminative Multitask Clustering (2013, background): “The second class shares common model parameters, such as placing a common prior across tasks within the hierarchical Bayesian framework [13], [14], [15], [16], learning the differences of the task-specific models in Frobenius norms under the regularization framework [17], [18], [19], etc.”
Recent Advances in Bayesian Optimization (2022, methodology): “Following the similar idea, a more general GP with parametric covariance functions is introduced by Lawrence and Platt [146] for MTL and knowledge sharing.”