Scalable Kernel Learning via the Discriminant Information
This work addresses scalability issues in kernel learning for machine learning practitioners, though it appears incremental as it builds on existing kernel approximation and discriminant analysis techniques.
The paper tackles the problem of high computational and memory complexity in kernel methods by developing a supervised kernel learning approach based on the Discriminant Information criterion to optimize kernel feature maps, resulting in improved optimization and generalization performances over state-of-the-art methods.
Kernel approximation methods create explicit, low-dimensional kernel feature maps to deal with the high computational and memory complexity of standard techniques. This work studies a supervised kernel learning methodology to optimize such mappings. We utilize the Discriminant Information criterion, a measure of class separability with a strong connection to Discriminant Analysis. By generalizing this measure to cover a wider range of kernel maps and learning settings, we develop scalable methods to learn kernel features with high discriminant power. Experimental results on several datasets showcase that our techniques can improve optimization and generalization performances over state of the art kernel learning methods.