Abstract
We describe our winning solution to the KDD Cup 2021 Open Benchmark Challenge. We tackle the task of academic paper classification within a heterogeneous graph containing paper, author and institution nodes. We present an efficient model based on our previously introduced algorithms: EMDE and Cleora, on top of a simplistic feed-forward neural network. Our solution can be trained on a single commodity 16 GB GPU, taking around 40 minutes per model.
Type
Publication
KDD Cup 2021 - OGB Large-Scale Challenge
Michal Daniluk
Research Scientist
My research interests include graph representation learning, recommendation systems, behavioral user representations, NLP.