Scribd
Upload
134 views1 page

CS 440: Final Exam, Question 1

1) The minimal expected cost C(S) of being in the starting cell S is the average of the cost of moving to its neighboring cells (1,0) and (0,1). 2) The minimal expected cost C((1,0)) of being in the neighboring cell (1,0) is the average of its costs to move to its own neighbors. 3) For any cell x, the minimal expected cost C(x) is a mathematical expression that averages the costs of moving from x to its neighboring cells, where each term represents the cost of moving to a particular neighbor.

Uploaded by

ChaituM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
134 views1 page

CS 440: Final Exam, Question 1

1) The minimal expected cost C(S) of being in the starting cell S is the average of the cost of moving to its neighboring cells (1,0) and (0,1). 2) The minimal expected cost C((1,0)) of being in the neighboring cell (1,0) is the average of its costs to move to its own neighbors. 3) For any cell x, the minimal expected cost C(x) is a mathematical expression that averages the costs of moving from x to its neighboring cells, where each term represents the cost of moving to a particular neighbor.

Uploaded by

ChaituM
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Computer Science Department - Rutgers University Spring 2021

CS 440: Final Exam, Question 1 16:198:440

Consider the following landscape, representing an icy mountain ridge.

You begin at square S and your goal is to get to square G. You can move up/down/left/right between squares.
However, because of the ice, each time you move in a given direction, there is a 0.1 probability you accidentally slide
in one of the orthogonal directions (if you try to go down, and you are not at the edge of the map, you successfully
go down 80% of the time, you slide left 10% of the time, and right 10% of the time). You cannot attempt to move
in the direction of a wall, and if one of the orthogonal directions is blocked by a wall you slide with a (90%, 10%)
split instead of (80%, 10%, 10%). You want to get to the goal as quickly as possible, as each timestep you are on
the map costs you one dollar.
But beware! The gray cells represent icy ravines, and if you enter one of those cells you slide to any icy death
(represented in this case by an immediate loss of $1000 and termination of the game).
Note: In general I adopt the notation that (x, y) is the cell in row x, column y.

1) (5 points) Let C(x) be the minimal expected cost of being in cell x. What is C(S) (in terms of its neighbors,
cell (1, 0) and (0, 1))?

2) (5 points) What is C((1, 0)) in terms of its neighbors?

2) (15 points) For any cell x, give a mathematical expression for C(x). Be clear on what each term or factor
represents.

3) (10 points) If you had the value of C(x) for each cell, how could you use it to find the best direction to move
in cell S?

4) (30 points) Compute C(x) for each x. What is C(S)?

5) (10 points) Determine for each cell what the best direction to move in is. Show this on the grid.

Bonus: (15 points) If the cost of the icy ravine death were $0, then you’d probably want to run to the nearest
ravine to terminate the game early and minimize your overall cost. Eventually, as you start increasing the cost of
falling down the ravine, you reach a point where it’s preferable to start trying to avoid the ravine. At what point (cost
threshold) do the best directions to take in each cell stop changing? How can you find that threshold, and what are
the best directions to move when above this threshold?

You might also like