HW1_Solutions

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

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: t) > CalcSum(dealer) and CalcSum(dealer) > 17 ... Action Hit Stand ... d) Using seed 0, simulating 50 rounds: Agent Type Percentage of Wins Probabilistic “target one” 40- 60% Probabilistic “target two” 40- 60% e) Answers to this question may vary. The utility based agent has an advantage is that the way it interacts with a model more naturally fits the blackjack problem. One particularly nice representation would be to use a utility function that uses the chance of winning after taking an action as the utility value. Part IV a) A goal- based agent. b) They may not format in a table like this, but their decisions should be similar. QuantSum = low...
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