Learning Shared Latent Structure for Image Synthesis and Robotic Imitation

By Aaron P. Shon, Keith Grochow, Aaron Hertzmann, Rajesh P. N. Rao. 2005

223 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

Adversarial Continual Learning (2020, background): “The approaches to tackle multi-view learning aim at either maximizing the mutual agreement on distinct views of the data or focus on obtaining a latent subspace shared by multiple views by assuming that the input views are generated from this latent subspace using Canonical correlation analysis and ”
Third-Person Imitation Learning (2017, background): “Such a mapping is often difﬁcult to obtain, and it typically relies on providing feature representations that captures the invariance between both environments Carpenter et al. (2002); Shon et al. (2005); Calinon et al. (2007); Nehaniv (2007); Gioioso et al. (2013); Gupta et al. (2016).”
A Survey on Multi-view Learning (2013, methodology): “Several methods (Shon et al., 2006; Sigal et al., 2009) have been designed to tackle this problem by constructing a latent subspace shared by multiple views, in which distinct views are connected with one another in this subspace, integrating the complementary information underlying different views.”
Bayesian group factor analysis with structured sparsity (2016, methodology): “Extensions of GPLVM include models that allow multiple observations (Shon et al., 2005; Ek et al., 2008; Salzmann et al., 2010; Damianou et al., 2012).”
Bayesian Optimization and Semiparametric Models with Applications to Assistive Technology (2014, methodology): “A GPLVM formulation in which multiple GPLVMs mapping to different signals share a single latent space, the shared GPLVM (SGPLVM), was introduced by Shon et al. [2005]. Wang et al.”
Non-Linear Domain Adaptation with Boosting (2013, background): “Shared latent variable models have also been proposed to learn a shared representation across multiple feature sources or tasks [4, 19, 6, 7, 21].”
Factorized Latent Spaces with Structured Sparsity (2010, background): “While the dictionaries are specific to each view, the weights of these dictionaries act as latent variables and are the same for all the views.”
Local distance preservation in the GP-LVM through back constraints (2006, unknown): “Indeed the successful applications and extensions of the GP-LVM (Grochow et al., 2004; Urtasun et al., 2005; Shon et al., 2006; Wang et al., 2006) each take explicit advantage of the fact that the model is probabilistic.”
Deep Variational Canonical Correlation Analysis (2016, methodology): “For VCCA/VCCA-private, we use Gaussian observation models as the inputs are real-valued.”
On Nonparametric Guidance for Learning Autoencoder Representations (2011, methodology): “Shon et al. (2005) introduced a Shared-GPLVM (SGPLVM) that used multiple GPs to map from a single shared latent space to various related signals. Wang et al. (2007) demonstrate that a generalisation of multilinear models arises as a GPLVM with product kernels, each mapping to different signals.”
Manifold Relevance Determination (2012, methodology): “Evolved shared models then ( left to right ) assume ﬁrstly, that all of the variance in the observations was shared (Shon et al., 2006).”
Multi-View Intact Space Learning (2015, methodology): “The proposed MISL algorithm and its kernel extension KMISL were compared with convex Multi-view Subspace Learning algorithm (MSL) [52], Factorized Latent Spaces with Structured Sparsity algorithm (FLSSS) [36], and shared Gaussian Latent Variable Model (sGPLVM) [33].”
Partial Multi-View Clustering (2014, background): “The former are extensions from single-view spectral clustering approaches with the help of some similarity measures (de Sa 2005; Zhou and Burges 2007; Kumar and Daumé 2011; Kumar, Rai, and Daumé 2011), whereas the latter generally try to identify a latent subspace across the multiple views (Hardoon ”
A Survey of Multi-View Representation Learning (2016, unknown): “[157] present the shared GPLVM (SGPLVM) model as a generalization of GPLVM that can handle multiple observation spaces, where each set of observations is parameterized by a different set of kernel parameters.”