Semiparametric latent factor models

By Yee Whye Teh, Matthias W. Seeger, Michael I. Jordan. 2005

273 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.

Citations in the corpus

Taking the Human Out of the Loop: A Review of Bayesian Optimization (2016, background): “[136] M.”
Deep Factors for Forecasting (2019, methodology): “By dropping the random effects, using GPs as the prior and removing the restriction of RNNs on the global factors, we recover the semi-parametric latent factor model (Teh et al., 2005).”
Input Warping for Bayesian Optimization of Non-Stationary Functions (2014, background): “However, a basic and surprisingly effective approach is to assume that each task is derived from a single latent function which is transformed to produce each output [Teh et al., 2005, Bonilla et al., 2008].”
Gaussian Process Regression Networks (2011, methodology): “In machine learning, the semiparametric latent factor model (SLFM) (Teh et al., 2005) was introduced to model multiple outputs with fixed signal correlations.”
Continual Multi-task Gaussian Processes (2019, methodology): “This model is also known in the literature as the semiparametric latent factor model (Teh et al., 2005).”
Latent Force Models (2009, background): “In (Teh et al., 2005) this is known as a semi-parametric latent factor model (SLFM), although their main focus is not the temporal case.”
Functional, but Not Anatomical, Separation of "what" and "when" in Prefrontal Cortex Gaussian-process Factor Analysis for Low-dimensional Single-trial Analysis of Neural Population Activity (unknown): “Relative to previously proposed models in the machine learning literature, GPFA is most similar to the semiparametric latent factor model (Teh et al. 2005), where the GPFA model can be obtained by letting time indices play the role of inputs.”
Sparse Convolved Gaussian Processes for Multi-output Regression (2008, methodology): “We have presented a sparse approximation for multiple output GPs, capturing the correlated information among outputs and reducing the amount of computational load for prediction and optimization purposes.”
Computationally Efficient Convolved Multiple Output Gaussian Processes (2011, methodology): “One of the paradigms that has been considered (Teh et al., 2005; Osborne et al., 2008; Bonilla et al., 2008) is known in the geostatistics literature asthe linear model of coregionalization(LMC) (Journel and Huijbregts, 1978; Goovaerts, 1997).”
Learning Treatment-Response Models from Multivariate Longitudinal Data (2017, methodology): “ucture which cannot be modeled by g i(t); this signal-speciﬁc structure is modeled by v id(t). The structure used in the random-effects component is similar to the linear models of coregionalization (Seeger et al., 2005; Alvarez and Lawrence, 2009), which is used´ to jointly model multiple correlate”
Unsupervised Learning of Image Transformations (2007, methodology): “[13]) our model can be trained on much larger datasets, and is capable of online learning.”
Multi-Task Learning using Generalized t Process (2010, background): “(Teh et al., 2005) proposed a semiparametric latent factor model for multi-output regression problems, a special case of multitask learning, in which the latent function values are modeled as linear combinations of multiple GP priors.”
Efficient Multioutput Gaussian Processes through Variational Inducing Kernels (2010, background): “Most attempts to apply such priors (Teh et al., 2005; Rogers et al., 2008; Bonilla et al., 2008) have focused on what is known in the geostatistics community as “linear model of coregionalization” (LMC) (Goovaerts, 1997).”
Bayesian Optimization for Synthetic Gene Design (2015, methodology): “In this work we use a combination of the linear and the intrinsic corregionalization models [16, 1].”
Multi-Task Bayesian Optimization (2013, background): “One simple approach is called the intrinsic model of coregionalization [12, 11, 13], which transforms a latent function to produce each output.”
Hidden Parameter Markov Decision Processes: A Semiparametric Regression Approach for Discovering Latent Task Parametrizations (2013, methodology): “Various optimization procedures exist for choosing the support points [15, 14]; we found that iteratively choosing support points from existing points to minimize the maximum reconstruction error within each batch was best for a setting in which a few large errors can result in poor performance.”
Spike and Slab Variational Inference for Multi-Task and Multiple Kernel Learning (2011, methodology): “Further, the model in (1) can be considered as a spike and slab Bayesian reformulation of multitask [14, 15] and multiple kernel learning previous methods [11, 12] that learn the weights using maximum likelihood.”
Multi-task Gaussian Process Learning of Robot Inverse Dynamics (2008, background): “An important related work is the semiparametric latent factor model [12] which has a number of latent processes which are linearly combined to produce observable functions as in eq.”