/MultiHopKG

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

Primary LanguageJupyter NotebookBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

Issues