The Recurrent Temporal Restricted Boltzmann Machine

By Ilya Sutskever, Geoffrey E. Hinton, Graham W. Taylor. 2008

447 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

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

In Probabilistic & Bayesian Methods

Citations in the corpus

Generating Sequences With Recurrent Neural Networks (2013, background): “Recurrent neural networks (RNNs) are a rich class of dynamic models that have been used to generate sequences in domains as diverse as music [6, 4], text [30] and motion capture data [29].”
Representation Learning: A Review and New Perspectives (2012, background): “For RBMs and some (not too deep) Boltzmann machines, one option is the use of Annealed Importance Sampling (Murray and Salakhutdinov, 2009) in order to estimate the partition function (and thus the test log-likelihood).”
Action-Conditional Video Prediction using Deep Networks in Atari Games (2015, methodology): “Thus, recent studies have mostly focused on modeling simple video data, such as bouncing balls or small patches, where the next frame is highly-predictable given the previous frames [29, 20, 19].”
Modeling Temporal Dependencies in High-Dimensional Sequences: Application to Polyphonic Music Generation and Transcription (2012, background): “…∂ ˆ h ( t ) ˆ h ( t ) (1 − ˆ h ( t ) ) (20) ∂C ∂ ˆ h ( t ) ˆ h ( t ) (1 − ˆ h ( t ) )ˆ h ( t − 1)T (21) 5 Baseline experiments In this section, we compare the performance of the RTRBM with the RNN-RBM on two baseline datasets: bouncing balls videos and motion capture data (Sutskever et al., 2008).”
Video (language) modeling: a baseline for generative models of natural videos (2014, methodology): “Bilinear models of transformations between nearby frames have also been investigated by Memisevic & Hinton (2009); Sutskever et al. (2009); Taylor et al. (2011); Michalski et al. (2014) as well as Miao & Rao (2007) via Lie group theory.”
Deep learning in neural networks: An overview (2014, neutral): “There are extensions for Temporal RBMs (Sutskever, Hinton, & Taylor, 2008).”
Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactions (2018, background): “Neural approaches that learn to directly model motion or physical interactions in pixel space offer an alternative solution (Srivastava et al., 2015; Sutskever et al., 2009).”
A Compositional Object-Based Approach to Learning Physical Dynamics (2016, background): “…over the parameters of a symbolic physics engine, while the bottom-up approach (Agrawal et al., 2016; Fragkiadaki et al., 2015b; Lerer et al., 2016; Li et al., 2016; Mottaghi et al., 2015; 2016; Sutskever et al., 2009) learns to directly map observations to motion prediction or physical judgments.”
Learning Visual Predictive Models of Physics for Playing Billiards (2015, methodology): “Video prediction Works of (Michalski et al., 2014; Sutskever et al., 2008) generate images of bouncing balls using a hierarchy of learnt image transformations in (Michalski et al.”
Structural-RNN: Deep Learning on Spatio-Temporal Graphs (2015, methodology): “Several approaches based on Gaussian processes [59, 63], Restricted Boltzmann Machines (RBMs) [57, 56, 51], and RNNs [14] have been proposed to model human motion.”
"What Happens If..." Learning to Predict the Effect of Forces in Images (2016, neutral): “[33] propose Recurrent RBMs to model high dimensional sequences.”
Learning Stochastic Recurrent Networks (2014, methodology): “Models based on restricted Boltzmann machines and variations [12, 13, 14] provide a solution to this as well, yet come with tighter restrictions on the assumptions that can be made.”
Learning to Generate Long-term Future via Hierarchical Prediction (2017, background): “Early work on future frame prediction focused on small patches containing simple predictable motions (Sutskever et al., 2009; Michalski et al., 2014; Mittelman et al., 2014) and motions in real videos (Ranzato et al., 2014; Srivas-tava et al., 2015).”
Temporal Generative Adversarial Nets with Singular Value Clipping (2016, background): “Temporal RBMs have also been proposed to explicitly capture the temporal correlations in videos [40, 38, 37].”
Short-Term Forecasting of Passenger Demand under On-Demand Ride Services: A Spatio-Temporal Deep Learning Approach (2017, result): “In the RNN architecture, the dependent variable in one timestamp was not only dependent on the explanatory variables in this timestamp, but also correlated with the explanatory variables in the previous timestamps (Graves, 2013; Sutskever et al., 2009).”
Pairwise input neural network for target-ligand interaction prediction (2014, methodology): “proposed pre-training and fine tuning procedure to train deep neural network to avoid the local optimum [13].”
Conditional Neural Fields (2009, background): “Very recently, the probabilistic neural language model [17] and recurrent temporal restricted Boltzmann machine [18] are proposed for natural language and time series modeling.”