CS 221 Study Guide - Final Guide: Perceptron, State Space, Unsupervised Learning

• School: Stanford University
• Department: Computer Science
• Course: CS 221
• Professor: all

Document Summary

The following pages are excerpts from similar classes" nals. The content is similar to what we"ve been covering this quarter, so that it should be useful for practicing. Note that the topics and terminology di er slightly, so feel free to ignore the questions that we did not cover. Certain topics are less emphasized in the past exams, but will be more emphasized in the nal for the class. These include: weighted csps and markov nets (the practice exams place more of an emphasis on. Bayes nets): loss-based learning (the practice exams place an emphasis on naive bayes instead), unsupervised learning (e. g. , em, logic (covered in much greater depth in our class) In contrast, the practice exams cover state space models fairly deeply. State space models will be less emphasized in the nal for the class. The rst portion of the practice exam comes with solutions; the rest are provided as example problems, but without solutions.