A Conditioned Unsupervised Regression Framework Attuned to the Dynamic Nature of Data Streams

arXiv:2312.07682v2 Announce Type: replace
Abstract: In scenarios where obtaining real-time labels proves challenging, conventional approaches may result in sub-optimal performance. This paper presents an optimal strategy for streaming contexts with limited labeled data, introducing an adaptive technique for unsupervised regression. The proposed method leverages a sparse set of initial labels and introduces an innovative drift detection mechanism to enable dynamic model adaptations in response to evolving patterns in the data. To enhance adaptability, we integrate the ADWIN (ADaptive WINdowing) algorithm …

abstract arxiv cs.lg data data streams dynamic framework labels nature paper performance real-time regression set strategy streaming type unsupervised