liusi2019 / ocd

Jupyter nodebook with R code for reproducing the experiments in the paper "Open Category Detection with PAC Guarantees". Need to use linux/windows for isolation forest part.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Open-Category-Detection-with-PAC-Guarantees

Jupyter nodebooks with R code for reproducing the experiments in the paper "Open Category Detection with PAC Guarantees".

Need to use linux/windows to do isolation forest part.

Need to install package btloda from liusi2019/btloda to do bootstrap and out-of-bag LODA part.

This repository is jointly contributed by Si Liu and Risheek Garrepalli.

If you find this useful in your research, please consider citing:

@inproceedings{liu2018open,
  title={Open Category Detection with PAC Guarantees},	
  author={Liu, Si and Garrepalli, Risheek and Dietterich, Thomas and Fern, Alan and Hendrycks, Dan},	
  booktitle={International Conference on Machine Learning},	
  pages={3175--3184},	
  year={2018}	
}

About

Jupyter nodebook with R code for reproducing the experiments in the paper "Open Category Detection with PAC Guarantees". Need to use linux/windows for isolation forest part.


Languages

Language:Jupyter Notebook 94.1%Language:R 5.9%