Generalized Denoising Auto-Encoders as Generative Models
نویسندگان
چکیده
Recent work has shown how denoising and contractive autoencoders implicitly capture the structure of the data-generating density, in the case where the corruption noise is Gaussian, the reconstruction error is the squared error, and the data is continuous-valued. This has led to various proposals for sampling from this implicitly learned density function, using Langevin and Metropolis-Hastings MCMC. However, it remained unclear how to connect the training procedure of regularized auto-encoders to the implicit estimation of the underlying datagenerating distribution when the data are discrete, or using other forms of corruption process and reconstruction errors. Another issue is the mathematical justification which is only valid in the limit of small corruption noise. We propose here a different attack on the problem, which deals with all these issues: arbitrary (but noisy enough) corruption, arbitrary reconstruction loss (seen as a log-likelihood), handling both discrete and continuous-valued variables, and removing the bias due to non-infinitesimal corruption noise (or non-infinitesimal contractive penalty).
منابع مشابه
GSNs : Generative Stochastic Networks
We introduce a novel training principle for generative probabilistic models that is an alternative to maximum likelihood. The proposed Generative Stochastic Networks (GSN) framework generalizes Denoising Auto-Encoders (DAE) and is based on learning the transition operator of a Markov chain whose stationary distribution estimates the data distribution. The transition distribution is a conditiona...
متن کاملCascading Denoising Auto-Encoder as a Deep Directed Generative Model
Recent work (Bengio et al., 2013) has shown how Denoising Auto-Encoders(DAE) become generative models as a density estimator. However, in practice, the framework suffers from a mixing problem in the MCMC sampling process and no direct method to estimate the test loglikelihood. We consider a directed model with an stochastic identity mapping (simple corruption process) as an inference model and ...
متن کاملGenerative Adversarial Source Separation
Generative source separation methods such as non-negative matrix factorization (NMF) or auto-encoders, rely on the assumption of an output probability density. Generative Adversarial Networks (GANs) can learn data distributions without needing a parametric assumption on the output density. We show on a speech source separation experiment that, a multilayer perceptron trained with a Wasserstein-...
متن کاملTowards Biologically Plausible Deep Learning
Neuroscientists have long criticised deep learning algorithms as incompatible with current knowledge of neurobiology. We explore more biologically plausible versions of deep representation learning, focusing here mostly on unsupervised learning but developing a learning mechanism that could account for supervised, unsupervised and reinforcement learning. The starting point is that the basic lea...
متن کاملLearning invariant features through local space contraction
We present in this paper a novel approach for training deterministic auto-encoders. We show that by adding a well chosen penalty term to the classical reconstruction cost function, we can achieve results that equal or surpass those attained by other regularized auto-encoders as well as denoising auto-encoders on a range of datasets. This penalty term corresponds to the Frobenius norm of the Jac...
متن کامل