# Bonus: Semi-supervied Learning Problem Description: * Remove some label in training data as **unlabeled data** * Implement semi-supervised learning algorithm such as self-training, clustering * Compare the performance with the model using only the labeled training data, record the training procedure ## Lecture <img src="semi.png" alt="Drawing" style="width: 800px;"/> ## Reference: → http://speech.ee.ntu.edu.tw/~tlkagk/courses/ML_2016/Lecture/semi%20(v3).pdf → [More Semi-supervised Learning Tutorial](http://pages.cs.wisc.edu/~jerryzhu/pub/sslicml07.pdf) → [上學期 Machine Learning 作業三的說明](http://speech.ee.ntu.edu.tw/~tlkagk/courses/ML_2016/Lecture/ML%20HW3.pdf)