Learning to Generate with Memory

Memory units have been widely used to enrich the capabilities of deep networks on capturing long-term dependencies in reasoning and prediction tasks, but little investigation exists on deep generative models (DGMs) which are good at inferring high-level invariant representations from unlabeled data. This paper presents a deep generative model with a possibly large external memory and an attention mechanism to capture the local detail information that is often lost in the bottom-up abstraction process in representation learning. By adopting a smooth attention model, the whole network is trained end-to-end by optimizing a variational bound of data likelihood via auto-encoding variational Bayesian methods, where an asymmetric recognition network is learnt jointly to infer high-level invariant representations. The asymmetric architecture can reduce the competition between bottom-up invariant feature extraction and top-down generation of instance details. Our experiments on several datasets demonstrate that memory can significantly boost the performance of DGMs on various tasks, including density estimation, image generation, and missing value imputation, and DGMs with memory can achieve state-of-the-art quantitative results.

[1]  Matthias Bethge,et al.  A note on the evaluation of generative models , 2015, ICLR.

[2]  Alex Graves,et al.  DRAW: A Recurrent Neural Network For Image Generation , 2015, ICML.

[3]  Simon Haykin,et al.  GradientBased Learning Applied to Document Recognition , 2001 .

[4]  Hugo Larochelle,et al.  The Neural Autoregressive Distribution Estimator , 2011, AISTATS.

[5]  Koray Kavukcuoglu,et al.  Visual Attention , 2020, Computational Models for Cognitive Vision.

[6]  Wojciech Zaremba,et al.  Reinforcement Learning Neural Turing Machines , 2015, ArXiv.

[7]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[8]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[9]  Yoshua Bengio,et al.  Show, Attend and Tell: Neural Image Caption Generation with Visual Attention , 2015, ICML.

[10]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[11]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[12]  Pascal Vincent,et al.  Generalized Denoising Auto-Encoders as Generative Models , 2013, NIPS.

[13]  Jason Weston,et al.  End-To-End Memory Networks , 2015, NIPS.

[14]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[15]  Pascal Vincent,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010, J. Mach. Learn. Res..

[16]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[17]  Max Welling,et al.  Semi-supervised Learning with Deep Generative Models , 2014, NIPS.

[18]  Zhe Gan,et al.  Learning Deep Sigmoid Belief Networks with Data Augmentation , 2015, AISTATS.

[19]  Bo Zhang,et al.  Learning Deep Generative Models with Doubly Stochastic MCMC , 2015, ArXiv.

[20]  Geoffrey E. Hinton,et al.  Deep Boltzmann Machines , 2009, AISTATS.

[21]  Ruslan Salakhutdinov,et al.  Evaluating probabilities under high-dimensional latent variable models , 2008, NIPS.

[22]  Bo Zhang,et al.  Max-Margin Deep Generative Models , 2015, NIPS.

[23]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[24]  Ryan P. Adams,et al.  Learning the Structure of Deep Sparse Graphical Models , 2009, AISTATS.

[25]  Nitish Srivastava,et al.  Learning Generative Models with Visual Attention , 2013, NIPS.

[26]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[27]  Wojciech Zaremba,et al.  Reinforcement Learning Neural Turing Machines - Revised , 2015 .

[28]  Geoffrey E. Hinton,et al.  Learning to combine foveal glimpses with a third-order Boltzmann machine , 2010, NIPS.

[29]  Tapani Raiko,et al.  Semi-supervised Learning with Ladder Networks , 2015, NIPS.

[30]  Yoshua Bengio,et al.  Deep Generative Stochastic Networks Trainable by Backprop , 2013, ICML.

[31]  Radford M. Neal.  Connectionist Learning of Belief Networks , 1992, Artif. Intell..

[32]  Alex Graves,et al.  Generating Sequences With Recurrent Neural Networks , 2013, ArXiv.

[33]  Daan Wierstra,et al.  Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.

[34]  Razvan Pascanu,et al.  Theano: new features and speed improvements , 2012, ArXiv.

[35]  Yoshua Bengio,et al.  Neural Machine Translation by Jointly Learning to Align and Translate , 2014, ICLR.

[36]  Daan Wierstra,et al.  Deep AutoRegressive Networks , 2013, ICML.

[37]  Ruslan Salakhutdinov,et al.  Importance Weighted Autoencoders , 2015, ICLR.

[38]  Rob Fergus,et al.  Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks , 2015, NIPS.

[39]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[40]  Jason Weston,et al.  Memory Networks , 2014, ICLR.

[41]  Yoshua Bengio,et al.  Reweighted Wake-Sleep , 2014, ICLR.

[42]  Harri Valpola,et al.  From neural PCA to deep unsupervised learning , 2014, ArXiv.

[43]  Phil Blunsom,et al.  Learning to Transduce with Unbounded Memory , 2015, NIPS.

[44]  Alex Graves,et al.  Neural Turing Machines , 2014, ArXiv.

[45]  Edward H. Adelson,et al.  The Laplacian Pyramid as a Compact Image Code , 1983, IEEE Trans. Commun..

[46]  Alex Graves,et al.  Recurrent Models of Visual Attention , 2014, NIPS.