Feature learning is a set of techniques used in machine learning to automatically discover representations from raw data for feature detection or classification. It replaces manual feature engineering and allows machines to learn the features and use them to perform a specific task. Feature learning can be supervised, unsupervised, or self-supervised.
UC Berkeley
Fall 2013
This course investigates the mathematical principles behind data and information analysis. It brings together concepts from statistics, optimization, and computer science, with a focus on large deviation inequalities, and convex analysis. It's tailored towards advanced graduate students who wish to incorporate these theories into their research.
No concepts data
+ 20 more concepts