English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Tree-structured LSTM is promising way to consider long-distance interactionover hierarchies. However, there have been few research efforts on thehyperparameter tuning of the construction and traversal of tree-structuredLSTM. To name a few, hyperparamters such as the interval of stateinitialization, the number of batches for normalization have been leftunexplored specifically in applying batch normalization for reducing trainingcost and parallelization. In this paper, we propose a novel recursive algorithmfor traversing batch normalized tree-structured LSTM. In proposal method, weimpose the constraint on the recursion algorithm for the depth-first search ofbinary tree representation of LSTM for which batch normalization is applied.With our constrained recursion, we can control the hyperparameter in thetraversal of several tree-structured LSTMs which is generated in the process ofbatch normalization. The tree traversal is divided into two steps. At firststage, the width-first search over models is applied for discover the startpoint of the latest tree-structured LSTM block. Then, the depth-first search isrun to traverse tree-structured LSTM. Proposed method enables us to explore theoptimized selection of hyperparameters of recursive neural networkimplementation by changing the constraints of our recursion algorithm. Inexperiment, we measure and plot the validation loss and computing time withchanging the length of internal of state initialization of tree-structuredLSTM. It has been turned out that proposal method is effective forhyperparameter tuning such as the number of batches and length of interval ofstate initialization of tree-structured LSTM.

A constrained recursion algorithm for batch normalization of tree-sturctured LSTM