qibinc / MultiHopKG

Multi-hop knowledge graph reasoning learned via policy gradient with reward shaping and action dropout

https://arxiv.org/abs/1808.10568

Geek Repo

Github PK Tool