Header menu link for other important links
FigureNet : A Deep Learning model for Question-Answering on Scientific Plots
Ameet Deshpande,
Published in Institute of Electrical and Electronics Engineers Inc.
Volume: 2019-July
Deep Learning has managed to push boundaries in a wide variety of tasks. One area of interest is to tackle problems in reasoning and understanding, with an aim to emulate human intelligence. In this work, we describe a deep learning model that addresses the reasoning task of question-answering on categorical plots. We introduce a novel architecture FigureNet, that learns to identify various plot elements, quantify the represented values and determine a relative ordering of these statistical values. We test our model on the FigureQA dataset which provides images and accompanying questions for scientific plots like bar graphs and pie charts, augmented with rich annotations. Our approach outperforms the state-of-the-art Relation Networks baseline by approximately 7% on this dataset, with a training time that is over an order of magnitude lesser. © 2019 IEEE.
About the journal
JournalData powered by TypesetProceedings of the International Joint Conference on Neural Networks
PublisherData powered by TypesetInstitute of Electrical and Electronics Engineers Inc.
Open AccessYes
Concepts (10)
  •  related image
    Statistical tests
  •  related image
    Area of interest
  •  related image
  •  related image
  •  related image
    Novel architecture
  •  related image
    Question answering
  •  related image
  •  related image
    State of the art
  •  related image
  •  related image
    Deep learning