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

In this paper, we present a method to mine object-part patterns fromconv-layers of a pre-trained convolutional neural network (CNN). The minedobject-part patterns are organized by an And-Or graph (AOG). This interpretableAOG representation consists of a four-layer semantic hierarchy, i.e., semanticparts, part templates, latent patterns, and neural units. The AOG associateseach object part with certain neural units in feature maps of conv-layers. TheAOG is constructed in a weakly-supervised manner, i.e., very few annotations(e.g., 3-20) of object parts are used to guide the learning of AOGs. We developa question-answering (QA) method that uses active human-computer communicationsto mine patterns from a pre-trained CNN, in order to incrementally explain morefeatures in conv-layers. During the learning process, our QA method uses thecurrent AOG for part localization. The QA method actively identifies objects,whose feature maps cannot be explained by the AOG. Then, our method asks peopleto annotate parts on the unexplained objects, and uses answers to discover CNNpatterns corresponding to the newly labeled parts. In this way, our methodgradually grows new branches and refines existing branches on the AOG tosemanticize CNN representations. In experiments, our method exhibited a highlearning efficiency. Our method used about 1/6-1/3 of the part annotations fortraining, but achieved similar or better part-localization performance thanfast-RCNN methods.

Mining Interpretable AOG Representations from Convolutional Networks via Active Question Answering