Robust multi-task learning with t-processes

By Shipeng Yu, Volker Tresp, Kai Yu. 2007

103 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, methodology): “When r equals 1, this model degenerates to a case where model parameters of different tasks share a prior, which is similar to several Bayesian MTL models such as [48], [49], [50] which are based on Gaussian processes and t processes.”
Learning Task Grouping and Overlap in Multi-task Learning (2012, neutral): “Other methods assume that there is one group of re-lated tasks and a small number of outlier tasks that are not related to any task in the pool (Yu et al., 2007; Chen et al., 2011).”
Recent Advances in Bayesian Optimization (2022, methodology): “As a treatment to the outlier tasks, Yu et al. [275] presented a robust extension of the previous studies [218, 274] by using heavier tailed t-Processes.”
Collaborating between Local and Global Learning for Distributed Online Multiple Tasks (2015, methodology): “MTL can be modeled by stochastic process methods, such as [29, 32].”
Semi-Supervised Multi-Task Regression (2009, methodology): “This problem also exists in the models of [16] and [17], which later motivated the development of a robust model using t-processes [23].”
Convex Multi-Task Learning by Clustering (2015, background): “There are numerous works on Bayesian models and algorithms for multi-task learning, mainly by sharing a prior, which can be used to relate tasks [13, 27].”
A convex formulation for learning shared structures from multiple tasks (2009, background): “…The problem of multi-task learning (Caruana, 1997) has recently received broad attention in areas such as machine learning, data mining, computer vision, and bioinformatics (Heisele et al., 2001; Ando & Zhang, 2005; Ando, 2007; Xue et al., 2007; Yu et al., 2007; Argyriou et al., 2008).”
Integrating low-rank and group-sparse structures for robust multi-task learning (2011, background): “Learning multiple tasks under this setting is usually referred to as robust multi-task learning [36].”
Multi-Task Learning using Generalized t Process (2010, methodology): “One of these is (Yu et al., 2007) as discussed above.”
Student-t Processes as Alternatives to Gaussian Processes (2014, methodology): “Despite the apparent ﬂexibility of the inverse Wishart distribution, we illustrate in Lemma 4 the surprising result that a multivariate Student-t distribution can be derived using a much simpler covariance prior which has been considered previously [Yu et al., 2007].”
Learning Multiple Tasks with a Sparse Matrix-Normal Penalty (2010, methodology): “These efforts include using mixtures of Gaussians [4] or Dirichlet processes [26] to model task groups, encouraging clustering of tasks v ia a convex regularization penalty [21], identifying “outlier” tasks by robust t-processes [29], an d i ferring task similarity from task-specific features [8, 27”
Multitask Compressive Sensing (2009, background): “borrowing of information, as some tasks can be outliers [45], [ 46 ].”
Multi-Task Learning with Gaussian Matrix Generalized Inverse Gaussian Model (2013, methodology): “…multi-task learning lie in discovering the relationship among the tasks either by directly learning the relatedness of the tasks (Xue et al., 2007; Yu et al., 2007; Jacob et al., 2008) or by mining the common feature structures shared by the tasks (Ando & Zhang, 2005; Zhang et al., 2005; Argyriou…”
On Multiplicative Multitask Feature Learning (2014, background): “One direction of these works directly learns the dependencies among tasks, either by modeling the correlated regression or classification noise (Greene, 2002), or assuming that the model parameters share a common prior (Yu et al., 2005; Lee et al., 2007; Xue et al., 2007; Yu et al., 2007; Jacob et a”
The spatial distribution in infinite dimensional spaces and related quantiles and depths (2014, background): “, [38]) on [0, 1] with mean m, degrees of freedom r ≥ 3 and covariance kernel K(t, s) = min(t, s).”