Flexible latent variable models for multi-task learning

By Jian Zhang, Zoubin Ghahramani, Yiming Yang. 2008

113 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.

In Artificial Intelligence

Citations in the corpus

A Survey on Multi-Task Learning (2017, background): “Moreover, in [47], [48], a latent variable model is proposed for”
Lifelong Learning for Sentiment Classification (2015, background): “Multi-task learning optimizes the learning of multiple related tasks at the same time (Caruana, 1997, Chen et al., 2011, Saha et al., 2011, Zhang et al., 2008).”
A Probabilistic Model for Dirty Multi-task Feature Selection (2015, methodology): “Multi-task feature selection methods are used to improve the process of inferring the model coefficients from the observed data under the sparsity assumption (Vogt & Roth, 2010; Hernández-Lobato et al., 2010; Obozinski et al., 2009; Xiong et al., 2007; Zhang et al., 2008).”
Graph-Structured Multi-task Regression and an Efficient Optimization Method for General Fused Lasso (2010, background): “In multi-task learning, we are interested in learning multiple related tasks jointly by analyzing data from all of the tasks at the same time instead of considering each task individually [21, 22, 15].”
Geometry preserving multi-task metric learning (2012, background): “To attack this problem, multi-task learning (MTL) has recently received considerable attention (Caruana 1997; Evgeniou and Pontil 2004; Argyriou et al. 2008; Argyriou and Evgeniou 2008; Zhang et al. 2008; Zhang and Yeung 2010a).”
Probabilistic Multi-Task Feature Selection (2010, unknown): “[30] proposed a latent variable model for multi-task learning by using the Laplace prior to enforce sparsity.”
Active Lifelong Learning With "Watchdog" (2018, unknown): “(Zhang, Ghahramani, and Yang 2008).”
Meta-Learning by Adjusting Priors Based on Extended PAC-Bayes Theory (2017, methodology): “As common in Hierarchal Bayesian methods (Blei et al., 2003; Zhang et al., 2008; Edwards & Storkey, 2016) we assume a hierarchical model with shared random variable ψ and task-speciﬁc random variables w i , i = 1 , ..., n (see Figure 5).”
Safe Policy Search for Lifelong Reinforcement Learning with Sublinear Regret (2015, background): “ously and share knowledge during the joint learning process (Wilson et al., 2007; Zhang et al., 2008).”
ELLA: An Efficient Lifelong Learning Algorithm (2013, methodology): “Several formulations of this idea have been proposed, including a probabilistic framework (Zhang et al., 2008) and a non-parametric Bayesian method that automatically selects the number of bases (Rai & Daumé III, 2010).”
Feature Selection (2016, methodology): “Hence, the 2,1-norm regularization is preferred and widely used in different scenarios such as multi-task learning (Obozinski et al. 2007; Zhang et al. 2008), anomaly detection (Li et al. 2017; Wu et al. 2017), and crowdsourcing (Zhou and He 2017).”
Score Function Features for Discriminative Learning: Matrix and Tensor Framework (2014, methodology): “Thus, our approach is applicable in many challenging practical scenarios.”
Active Multitask Learning Using Both Latent and Supervised Shared Topics (2014, methodology): “These include different learning methods, such as empirical risk minimization using group-sparse regularizers [20, 17], hierarchical Bayesian models [43, 23] and hidden conditional random fields [29].”
Computational Sociolinguistics: A Survey (2016, background): “As another example, latent variable modeling approaches [143] have grown in prominence within the CL community for dimensionality reduction, managing heterogeneity in terms of multiple domains or multiple tasks [278], and approximation of semantics [25, 100].”
Learning Multiple Tasks using Manifold Regularization (2010, background): “There is also a large body of work on multitask learning that find the shared structure in the tasks using Bayesian inference [23] [24] [9], which in spirit, is similar to the above approaches, but done in a Bayesian way.”
Metalearning: a survey of trends and technologies (2013, background): “[70] approaches the subject from a Bayesian perspective, using a framework including latent variables for modelling a shared structure among different learning scenarios.”
Online Multi-Task Learning for Policy Gradient Methods (2014, methodology): “When data is in limited supply, learning task models jointly through multi-task learning (MTL) rather than independently can significantly improve model performance (Thrun & O'Sullivan, 1996; Zhang et al., 2008; Rai & Daumé, 2010; Kumar & Daumé, 2012).”
Multi-Task Feature Learning Via Efficient l2, 1-Norm Minimization (2009, methodology): “[28] proposed a unified probabilistic framework, where the task parameters share a common structure through latent variables.”