CDSeer: A Novel Semi-Supervised Concept Drift Detection Technique for Machine Learning Systems in Industrial Settings
This paper introduces CDSeer, a novel semi-supervised concept drift detection technique designed to address the limitations of existing methods in industrial settings, particularly regarding excessive labeling effort, inflexibility in labeling, and lack of generality across different machine learning models.
지속 엔트로피 및 토폴로지 보존 기반의 비지도 학습 방식의 데이터 변화 감 assessment
본 논문에서는 데이터 스트림의 토폴로지적 특징 변화를 감지하여 개념 변화를 식별하는 새로운 비지도 학습 프레임워크를 제안합니다.
Unsupervised Detection of Topological Changes in Data Streams Using Persistent Entropy and Dimensionality Reduction
This paper proposes a novel concept drift detection framework that goes beyond statistical changes by incorporating topological data analysis to identify significant shifts in the topological features of streaming data.
Evaluating the Tradeoff Between Accuracy and Energy Efficiency in Concept Drift Detection Methods for Sustainable Machine Learning Systems
There is no single concept drift detection method that excels across both accuracy and energy efficiency metrics. Practitioners must carefully weigh the tradeoffs to select the most suitable method for their ML-enabled systems.
© 2024 by Linnk AI