TL;DR通过观察后向传播错误梯度以及隐变量的激活,我们探讨了用 Langevin MCMC 方法估计基于能量的模型中的潜变量。我们提出了连续变量潜变量理论与 BP 算法在深层网络中的学习方式高效类比存在的可能性。
Abstract
We show that langevin mcmc inference in an energy-based model with latent
variables has the property that the early steps of inference, starting from a
stationary point, correspond to propagating error gradients