Deep Generative Networks (DGNs) with probabilistic modeling of their output and latent space are currently trained via Variational Autoencoders (VAEs). In the absence of a known analytical form for the posterior and likelihood expectation, VAEs resort to approximations, including (Amortized) Variational Inference (AVI) and Monte-Carlo (MC) sampling. We exploit the Continuous Piecewise Affine (CPA) property of modern DGNs to derive their posterior and marginal distributions as well as the latter's first moments. These findings enable us to derive an analytical Expectation-Maximization (EM) algorithm that enables gradient-free DGN learning. We demonstrate empirically that EM training of DGNs produces greater likelihood than VAE training. Our findings will guide the design of new VAE AVI that better approximate the true posterior and open avenues to apply standard statistical tools for model comparison, anomaly detection, and missing data imputation.
翻译:目前,通过变异自动计算器(VAE)培训了具有输出和潜在空间概率模型的深生成网络(DGNs),在没有已知的后向和概率预期分析表的情况下,VAEs采用近似法,包括(模拟)变位推理(AVI)和Monte-Carlo(MC)取样。我们利用现代DGs的连续小结(CPA)特性来得出其后向和边缘分布以及后者的最初时刻。这些发现使我们能够得出分析性预期-最大化(EM)算法,使无梯度DGN学习成为可能。我们从经验上表明,对DGNS的EM培训比VAE培训更有可能。我们的调查结果将指导新的VAE AVI的设计,使之更接近真实的后向和开放的路径,以应用标准统计工具进行模型比较、异常检测和缺失的数据干扰。