arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1306.1091cs.LG
收藏

可通过反向传播训练的深度生成随机网络

Deep Generative Stochastic Networks Trainable by Backprop

Université de Montréal(蒙特利尔大学);Cornell University(康奈尔大学)

Yoshua Bengio, Éric Thibodeau-Laufer, Guillaume Alain, Jason Yosinski

摘要:

我们引入了一种新的概率模型训练原理,它是最大似然的替代方案。提出的生成随机网络(GSN)框架是基于学习马尔可夫链的转移算子,其平稳分布估计数据分布。马尔可夫链的转移分布是以前一个状态为条件的,通常涉及一个小的移动,所以这个条件分布具有较少的主导模式,在小移动的极限下是单峰的。因此,它更容易学习,因为它更容易近似其分区函数,更像是学习执行监督函数近似,可以通过反向传播获得梯度。我们提供的定理,概括了最近的工作去噪自动编码器的概率解释,并获得了一个有趣的理由依赖网络和广义pseudolikketrium,随着一个适当的联合分布和采样机制的定义,即使当条件不一致。GSN可用于缺失输入,并可用于在给定其余变量的情况下对变量子集进行采样。我们使用模拟深度玻尔兹曼机吉布斯采样器的架构,在两个图像数据集上进行实验,验证了这些理论结果,但允许通过简单的反向传播进行训练,而不需要分层预训练。

英文摘要:

We introduce a novel training principle for probabilistic models that is an alternative to maximum likelihood. The proposed Generative Stochastic Networks (GSN) framework is based on learning the transition operator of a Markov chain whose stationary distribution estimates the data distribution. The transition distribution of the Markov chain is conditional on the previous state, generally involving a small move, so this conditional distribution has fewer dominant modes, being unimodal in the limit of small moves. Thus, it is easier to learn because it is easier to approximate its partition function, more like learning to perform supervised function approximation, with gradients that can be obtained by backprop. We provide theorems that generalize recent work on the probabilistic interpretation of denoising autoencoders and obtain along the way an interesting justification for dependency networks and generalized pseudolikelihood, along with a definition of an appropriate joint distribution and sampling mechanism even when the conditionals are not consistent. GSNs can be used with missing inputs and can be used to sample subsets of variables given the rest. We validate these theoretical results with experiments on two image datasets using an architecture that mimics the Deep Boltzmann Machine Gibbs sampler but allows training to proceed with simple backprop, without the need for layerwise pretraining.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑