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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Detecting statistical interactions between input features is a crucial andchallenging task. Recent advances demonstrate that it is possible to extractlearned interactions from trained neural networks. It has also been observedthat, in neural networks, any interacting features must follow a stronglyweighted connection to common hidden units. Motivated by the observation, inthis paper, we propose to investigate the interaction detection problem from anovel topological perspective by analyzing the connectivity in neural networks.Specially, we propose a new measure for quantifying interaction strength, basedupon the well-received theory of persistent homology. Based on this measure, aPersistence Interaction detection~(PID) algorithm is developed to efficientlydetect interactions. Our proposed algorithm is evaluated across a number ofinteraction detection tasks on several synthetic and real world datasets withdifferent hyperparameters. Experimental results validate that the PID algorithmoutperforms the state-of-the-art baselines.

Towards Interaction Detection Using Topological Analysis on Neural Networks