Factored conditional restricted Boltzmann Machines for modeling motion style

By Graham W. Taylor, Geoffrey E. Hinton. 2009

438 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

In Generative Models

Citations in the corpus

Generating Sequences With Recurrent Neural Networks (2013, background): “One remedy that has been proposed for conditional models is to inject noise into the predictions before feeding them back into the model [31], thereby increasing the model’s robustness to surprising inputs.”
Representation Learning: A Review and New Perspectives (2012, methodology): “It has been used in a truly impressive variety of applications, including fMRI image classification [180], motion and spatial transformations [201], [144], collaborative filtering [178], and natural image modeling [160], [53].”
Action-Conditional Video Prediction using Deep Networks in Atari Games (2015, methodology): “The advantages of multiplicative interactions have been explored in image and text processing [33, 30, 18].”
Dynamic Filter Networks (2016, background): “Multiplicative interactions in neural networks have also been investigated by [21].”
Structural-RNN: Deep Learning on Spatio-Temporal Graphs (2015, background): “Such modularity is of interest and has been explored to generate diverse motion styles [55].”
Emergence of Complex-Like Cells in a Temporal Product Network with Local Receptive Fields (2010, background): “Hinton and his collaborators have proposed several temporal models that use 3-way multiplicative interactions (Brown and Hinton, 2001; Taylor and Hinton, 2009).”
Deep Learning of Representations: Looking Forward (2013, methodology): “Since the parameters of the RBM are obtained through a parametric computation taking x as input,21 this is really a conditional RBM (Taylor et al., 2007; Taylor and Hinton, 2009).”
Semantic Compositional Networks for Visual Captioning (2016, neutral): “Similar ideas have been exploited in [25, 30, 38, 39, 41, 50, 14].”
Learning to Relate Images (2011, neutral): “22.”
A Multiplicative Model for Learning Distributed Text-Based Attribute Representations (2014, methodology): “When our factorization is embedded into a neural language model, it allows us to generate text conditioned on different attributes in the same manner as [11] could generate motions from different styles.”
Factored 3-Way Restricted Boltzmann Machines For Modeling Natural Images (2010, background): “In conditional 3-way factored models (Taylor and Hinton, 2009, Memisevic and Hinton, 2010), one of the two sets of visible variables is held constant during reconstruction.”
Training Restricted Boltzmann Machines (2015, background): “, 2009; Larochelle and Bengio, 2008); learning movement patterns (Taylor et al., 2007; Taylor and Hinton, 2009); collaborative filtering for movie recommendations (Salakhutdinov et al.”
Training restricted Boltzmann machines: An introduction (2014, background): “1 Introduction In the last years, models extending or borrowing concepts from restricted Boltzmann machines (RBMs, [49]) have enjoyed much popularity for pattern analysis and generation, with applications including image classification, processing, and generation [25, 51, 34, 28, 48, 31]; learning m”
Shallow vs. Deep Sum-Product Networks (2011, background): “Sin ce then, these ideas have been investigated further and applied in many settings, demonstratin g state-of-the-art learning performance in object recognition [16, 28, 18, 15] and segmentation [20] , audio classification [19, 10], natural language processing [9, 36, 21, 32], collaborative filterin”
Video Relationship Reasoning Using Gated Spatio-Temporal Energy Graph (2019, methodology): “Thanks to the recent advances in graphical models [33, 34, 21], deep neural networks [22, 35], and tensor factorization [14], our workaround is to parametrize and approximate f as f θ with learnable parameters θ as follows:”