CoPER: Contextual Parameter Generation for Knowledge Graph Link Prediction

This repository contains the official implementation of

Contextual Parameter Generation for Knowledge Graph Link Prediction
George Stoica*, Otilia Stretcu*, Emmanouil Antonios Platanios*, Tom Mitchell, and Barnabás Póczos
In Proceedings of the Thirty-fourth AAAI Conference on Artificial Intelligence 2020

For details on this work please check out our:

• AAAI 2020 paper
• supplementary material
• poster
• NeurIPS 2019 Graph Representation Workshop Paper

Code Structure

This repository contains our source code for both CoPER-ConvE and CoPER-MINERVA models, as well as supplementary material.

For details on running each method and their corresponding baselines (ConvE and Minerva), please navigate to the CoPER-ConvE and CoPER-MINERVA directories.

Citation

If using this code, please cite our work as:

@inproceedings{stoica2020coper,
  title     = {Contextual Parameter Generation for Knowledge Graph Link Prediction.},
  author    = {Stoica*, George and Stretcu*, Otilia and Platanios*, Emmanouil Antonios and Mitchell, Tom M and P{\'o}czos, Barnab{\'a}s},
  booktitle = {Proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence},
  pages     = {3000--3008},
  year      = {2020}
}