In this chapter, we've discussed the various ways we can categorize machine learning techniques. In particular, we discussed the difference between unsupervised learning, supervised learning, and reinforcement learning, presenting various examples of each.
We also discussed different ways to judge the accuracy of machine learning algorithms, in particular, the concepts of accuracy, precision, and recall as applied to supervised learning techniques. We also discussed the importance of the training step in supervised learning algorithms, and illustrated the concepts of bias, variance, generalization, and overfitting.
Finally, we looked at how machine learning algorithms can be categorized not by learning mode but instead by task or technique, and presented a number of algorithms that fit into the categories of clustering, classification, regression, dimensionality reduction...
