HW1_Solutions

# Think of a long narrow corridor the randomized agent

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: omized agent will have trouble navigating through this corridor will often change directions, etc. (e) Yes, a reﬂex agent with state can outperform a simple reﬂex agent because it can remember where it has patrolled already and look for new areas to clean. 3. (a) In order to behave rationally, the agent must sense whether the Suck action has been successful after it has been performed. If the square is still dirty, it must Suck again. (b) In the environment with a puppy, a rational agent will remember how long it has been since it has sensed each square is clean. After a period of time, the agent can go back and make sure the square is still clean. In this case, the agent must keep touring the squares indeﬁnitely. The probability that a square is dirty increases monotonically with the time since it was last cleaned, so the rational strategy is, roughly speaking, to repeatedly execute the shortest possible tour of all squares. 3 Exercise 4 Part I a)...
View Full Document

## This note was uploaded on 01/22/2014 for the course EECS 492 taught by Professor Staff during the Fall '08 term at University of Michigan.

Ask a homework question - tutors are online