TL;DRAutoregressive Energy Machine 是一种高效的基于能量的神经网络模型,具备在无监督学习中广泛使用的灵活性,可以在不受概率密度限制的条件下计算归一化常数,实现在密度估计任务中的最优表现。
Abstract
neural density estimators are flexible families of parametric models which
have seen widespread use in unsupervised machine learning in recent years.
Maximum-likelihood training typically dictates that these mode