Machine Learning

This course provides an introduction to computational mechanisms that improve their performance based on experience. Machine learning can be used in engineered systems for a wide variety of tasks in personalized information filtering, health care, security, games, computer vision, and human-computer interaction, and can provide computational models of information processing in biological and other complex systems. Supervised and unsupervised learning will be discussed, including sample applications, as well as specific learning paradigms such as decision trees, instance-based learning, neural networks and deep learning, Bayesian approaches, meta-learning, and clustering. General concepts to be described include feature space representations, inductive bias, overfitting, and fundamental tradeoffs.

Averages use numerical 1–5 ratings, including the ratings submitted with new anonymous reviews.

OrganizationHow well the course was organized

4.03 / 5

ChallengeHow intellectually challenging students found it

4.69 / 5

AttendanceHow necessary attendance was

4.32 / 5

AssignmentsHow helpful assignments were

4.09 / 5