论文标题
胶囊网络作为生成模型
Capsule Networks as Generative Models
论文作者
论文摘要
胶囊网络是一种专门用于视觉场景识别的神经网络体系结构。功能和姿势信息是从场景中提取的，然后通过称为“胶囊”的矢量值节点的层次结构进行动态路由，以创建隐式场景图，其最终目的是将视觉直接作为逆图。但是，尽管有这些直觉，但胶囊网络并未作为显式概率生成模型表达。此外，通常使用的路由算法是临时的，主要是由算法直觉动机。在本文中，我们得出了一种替代的胶囊路由算法，利用稀疏性约束下的迭代推断。然后，我们基于变压器网络中的自我发项操作引入了胶囊网络的显式概率生成模型，并使用von-mises-fisher（vmf）圆形高斯分布显示了它与预测性编码网络的变体的相关性。
Capsule networks are a neural network architecture specialized for visual scene recognition. Features and pose information are extracted from a scene and then dynamically routed through a hierarchy of vector-valued nodes called 'capsules' to create an implicit scene graph, with the ultimate aim of learning vision directly as inverse graphics. Despite these intuitions, however, capsule networks are not formulated as explicit probabilistic generative models; moreover, the routing algorithms typically used are ad-hoc and primarily motivated by algorithmic intuition. In this paper, we derive an alternative capsule routing algorithm utilizing iterative inference under sparsity constraints. We then introduce an explicit probabilistic generative model for capsule networks based on the self-attention operation in transformer networks and show how it is related to a variant of predictive coding networks using Von-Mises-Fisher (VMF) circular Gaussian distributions.
