6.01 Lecture 1: 6.01 Course Notes April 1, 2015 (Probability)

• School: MIT
• Department: Electrical Engineering and Computer Science
• Course: 6.01
• Professor: Adam Hartz

Document Summary

We can make observations of our local surroundings, We know what actions we have taken and the consequences those are likely to have on our location. We"ll assume that, as you navigate, you maintain a belief state which contains your best information about what state you"re in, which is represented as a probability distribution over all possible states. We"ll start by de ning probability distributions on simple and complex spaces, and develop a set of methods for de ning and manipulating them. Then, we"ll formulate problems like the navigation problem discussed above as state estimation problems in stochastic state machines. Finally, we"ll develop an algorithm for estimating the unobservable state of system, based on observations of its outputs. We have been using state machines to model systems and how they change over time. The state of a system is a description of the aspects of the system that allow us to predict its behavior over time.