Explores non-linear SVM using kernels for data separation in higher-dimensional spaces, optimizing training with kernels to avoid explicit transformations.
Discusses kernel methods in machine learning, focusing on kernel regression and support vector machines, including their formulations and applications.
Explores learning the kernel function in convex optimization, focusing on predicting outputs using a linear classifier and selecting optimal kernel functions through cross-validation.
Explores atomic descriptors, emphasizing symmetry, locality, and the challenges of incorporating electrostatics in machine learning models for chemistry.