PROJECT TITLE :
Sufficient Canonical Correlation Analysis
Canonical correlation analysis (CCA) is a good method to find two acceptable subspaces in that Pearson’s correlation coefficients are maximized between projected random vectors. Thanks to its well-established theoretical support and comparatively economical computation, CCA is widely used as a joint dimension reduction tool and has been successfully applied to many image processing and laptop vision tasks. However, as reported, the ancient CCA suffers from overfitting in many practical cases. During this paper, we tend to propose sufficient CCA (S-CCA) to alleviate CCA’s overfitting drawback, which is impressed by the idea of sufficient dimension reduction. The effectiveness of S-CCA is verified both theoretically and experimentally. Experimental results conjointly demonstrate that our S-CCA outperforms some of CCA’s fashionable extensions throughout the prediction section, especially when severe overfitting occurs.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here