Incremental ELMVIS for unsupervised learning
This incremental approach addresses scalability issues in unsupervised learning for researchers and practitioners dealing with large datasets, though it is an incremental improvement over existing methods.
The authors tackled the problem of scaling unsupervised learning by proposing an incremental version of ELMVIS+ that iteratively selects and adds the best-fitting samples from a large pool, reducing memory requirements while maintaining high speed. The method showed promising results in learning dependencies from non-organized data, such as reconstructing shuffled datasets or handling complex high-dimensional spaces.
An incremental version of the ELMVIS+ method is proposed in this paper. It iteratively selects a few best fitting data samples from a large pool, and adds them to the model. The method keeps high speed of ELMVIS+ while allowing for much larger possible sample pools due to lower memory requirements. The extension is useful for reaching a better local optimum with greedy optimization of ELMVIS, and the data structure can be specified in semi-supervised optimization. The major new application of incremental ELMVIS is not to visualization, but to a general dataset processing. The method is capable of learning dependencies from non-organized unsupervised data -- either reconstructing a shuffled dataset, or learning dependencies in complex high-dimensional space. The results are interesting and promising, although there is space for improvements.