Machine Learning (CSCI 5525)
Jan 27th, 2020
Email: Well, I am bad at emails. Not the best way to reach me. Include the substring “CSCI 5525” to begin a meaningful subject line.
Please post questions about course material and homework assignments on Canvas first, and then use emails only after an appropriate amount of time has passed without a response. Please use your UMN email account.
Labeled examples: (x_1, y_1), (x_2, y_2), \ldots (x_n, y_n)
each x_i is a feature vector (or representation) of an instance (e.g. image, audio, medical record)
each y_i is a task-specific label (e.g. cats versus dogs images, male versus female voices, risk of lung cancer)
Goal: learn a predictor \hat f\colon X \rightarrow Y based on labeled examples, that accurately predicts the labels of future x.
Turning weak learners into strong learners.