Random Cut Forest

Random Cut Forest (RCF) algorithms are a class of machine learning methods primarily used for anomaly detection, particularly in high-dimensional datasets and time series. Research focuses on improving RCF's efficiency and robustness, including variations like Robust RCF (RRCF) and weighted RCF (WRCF), which employ adaptive or density-based splitting strategies to enhance performance compared to simpler methods like Isolation Forest. These advancements are significant because effective anomaly detection is crucial for various applications, from identifying outliers in financial transactions to detecting faults in complex systems.

Papers

March 14, 2023

Systematic design space exploration by learning the explored space using Machine Learning
Avinash Kumar, Anish Kumar, Sumit Sharma, Surjeet Singh, Kumar Vardhan
Machine Learning Deep Space Geometric Learning Euclidean Space Parameter Space Exploration Random Cut Forest

June 18, 2022

Reduced Robust Random Cut Forest for Out-Of-Distribution detection in machine learning models
Harsh Vardhan, Janos Sztipanovits
Machine Learning Model Distribution Detection Distribution Input Random Cut Forest

February 1, 2022

Weighted Isolation and Random Cut Forest Algorithms for Anomaly Detection
Sijin Yeom, Jae-Hun Jung
Anomaly Detection Social Isolation Random Cut Forest

Random Cut Forest

Papers

Systematic design space exploration by learning the explored space using Machine Learning

Reduced Robust Random Cut Forest for Out-Of-Distribution detection in machine learning models

Weighted Isolation and Random Cut Forest Algorithms for Anomaly Detection