Streaming View Learning
This addresses a practical issue in multi-view learning for applications where data collection leads to streaming views, offering an incremental improvement over conventional methods.
The paper tackles the problem of multi-view learning when views arrive sequentially rather than simultaneously, proposing an algorithm that fine-tunes combination weights of stable subspaces from past views to accommodate new views, reducing the need to learn new view functions or update past ones, with experimental results on real-world datasets showing its effectiveness.
An underlying assumption in conventional multi-view learning algorithms is that all views can be simultaneously accessed. However, due to various factors when collecting and pre-processing data from different views, the streaming view setting, in which views arrive in a streaming manner, is becoming more common. By assuming that the subspaces of a multi-view model trained over past views are stable, here we fine tune their combination weights such that the well-trained multi-view model is compatible with new views. This largely overcomes the burden of learning new view functions and updating past view functions. We theoretically examine convergence issues and the influence of streaming views in the proposed algorithm. Experimental results on real-world datasets suggest that studying the streaming views problem in multi-view learning is significant and that the proposed algorithm can effectively handle streaming views in different applications.