Hierarchical Adaptive Pooling by Capturing High-Order Dependency for Graph Representation Learning

High-order Graph-based Neural Dependency Parsing

In this work, we present a novel way of using neural network for graph-based dependency parsing, which fits the neural network into a simple probabilistic model and can be furthermore generalized to high-order parsing. Instead of the sparse features used in traditional methods, we utilize distributed dense feature representations for neural network, which give better feature representations. Th...

Differentiable Pooling for Hierarchical Feature Learning

We introduce a parametric form of pooling, based on a Gaussian, which can be optimized alongside the features in a single global objective function. By contrast, existing pooling schemes are based on heuristics (e.g. local maximum) and have no clear link to the cost function of the model. Furthermore, the variables of the Gaussian explicitly store location information, distinct from the appeara...

Capturing Temporal Sequences for Text Representation Using Hierarchical SOMs

This paper explores a new representation for text document by using an unsupervised learning mechanism – a hierarchical structure of Self-Organizing Feature Maps. This architecture is based on a code-book model. Through this architecture, a document can be encoded to a sequence of neurons while the temporal sequences of words as well as their frequencies are kept. Our results show that this arc...

Video Representation Learning Using Discriminative Pooling

Popular deep models for action recognition in videos generate independent predictions for short clips, which are then pooled heuristically to assign an action label to the full video segment. As not all frames may characterize the underlying action—indeed, many are common across multiple actions—pooling schemes that impose equal importance on all frames might be unfavorable. In an attempt to ta...

High-Order Shock Capturing for Computational Aeroacoustics