online learning

Statistical Inference for Streamed Longitudinal Data

Parallel-and-Stream Accelerator for Computationally Fast Supervised Learning


MORA is a real-time regression analysis method that incorporates both dynamics and inter-data batch correlation.


A hybrid paradigm that integrates online streaming processing into each parallelized data process in a MapReduce framework.

Renewable Estimation and Incremental Inference in Generalized Linear Models with Streaming Datasets

This paper presents an incremental updating algorithm to analyze streaming datasets using generalized linear models. The proposed method is formulated within a new framework of renewable estimation and incremental inference, in which the estimates …


A new framework of real-time estimation and incremental inference in generalized linear models (GLMs) with cross-sectional data streams.