AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Kernelized Hashcode Representations for Relation Extraction

Sahil Garg, Aram Galstyan, Greg Ver Steeg, Irina Rish, Guillermo Cecchi, Shuyang Gao

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use KLSH-RF instead of the strongest reported baselineBiomedical relation extraction · AIMed → BioInferF1: 0.57 -> 0.57No clear effect
Use KLSH-RF instead of RNN or KLSH-kNNBiomedical relation extraction · BioInfer → AIMedF1: 0.51 -> 0.54Improved
Use KLSH-RF instead of CNNBiomedical relation extraction · PubMed45F1: 0.51 ± 0.28 -> 0.57 ± 0.25Improved
Use KLSH-RF instead of Bi-LSTMBiomedical relation extraction · PubMed45-ERNF1: 0.37 ± 0.15 -> 0.45 ± 0.22Improved
Use KLSH-RF instead of CNN or LSTM-CNNBiomedical relation extraction · BioNLPF1: 0.60 -> 0.63Improved
Optimize the reference set instead of selecting it randomly (ablation)KLSH-RF · PubMed45recall: 7-13 ptsImproved
Optimize the reference set instead of selecting it randomly (ablation)KLSH-RF · PubMed45precision: 2-3 ptsMixed
Optimize nonstationary kernel parameters (ablation)KLSH-RF with PK · PubMed45recall: 7 ptsImproved
Optimize nonstationary kernel parameters (ablation)KLSH-RF with PK · PubMed45precision: 1 ptHurt
Use fewer hashcode bits (H=15 or 30) (ablation)KLSH-RF · PubMed45F1Hurt
Increase the number of decision trees beyond a minimal threshold (ablation)KLSH-RF · PubMed45accuracyNo clear effect
Use Kulis hashing with higher alpha (ablation)KLSH-RF · PubMed45accuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy