Suppose you wanted to write a minimax-based Connect Four (https://en.wikipedia.o

Question

Suppose you wanted to write a minimax-based Connect Four (https://en.wikipedia.org/wiki/Connect_Four) playing agent. 1. How would you represent the "game tree" for your agent? 2. If you have infinite look-ahead, what heuristic function would you use to evaluate leaf nodes? 3. Here’s an example of a heuristic (let’s call it "The Simple Heuristic") you might use if you could only look ahead a finite number of plies (e.g., 6): •+100 for a winning leaf •-100 for a losing leaf •+5 for each of your discs that are in a three-in-a-row •-5 for each of your opponent’s discs that are in a three-in-a-row Define a heuristic of your own, and justify why it’s better than this heuristic.

Explanation / Answer

Ans:- (1). In base of the minimax-based Connect we can represent the game tree with the help of minimax-based Connect . First of all we are represent the rating of function. suppose our rating-function analyzes by playing area, so that the rates those player presently appears to have good chance of won the game, & other side outputs its positive as well as negative number. we can calculate the position of rating by inspection of each placed any kind of chip & seen number of possible lines of four. so the line of four still possible to make complete with the chip & totally complete already. When the chip likely to belongs to the other position (eg Al) so we can add points of all possible line. However chip related to belongs to the player of game the similar amount of points get. so we are a complete four-line will be rating of + or - it only depending on the player.

(2).Our infinite look-ahead is use to evaluate leaf nodes of a configuration any king of game. if you like or Unlike in heuristic function where evaluation leaf node function always a may me positive or negative so we are calculating the of the main cost for start evaluation node a goal and passing to given main node, so we are evaluation node function some time we called heuristic evaluation node function .

for example Evaluation node Function in heuristic function     

Suppose that for Sic /Sac/Soe:
f(n) = [ 3-lengths enter or open] - [3-lengths enter or open]
So that a 3-length is a fully complete in a row as well as column & diagonal.

(3). the heuristic static evaluation function to be applied at the Simple Heuristic leaf nodes, First of all we are calculating the main backed-up values at main parents leaf. so our node A is a -100 MIN node related to losing leaf main fact our position are opponent discs change to turn move in three-in-a-row . so value is that -1 (= min(100, 3, -1),suppose our node A is a +100 MAX node related to losing leaf main fact our position are opponent discs change to turn move in three-in-a-row . so value is that +1 (= max(100, 3, +1),suppose our values to the next upper level +5 of each discs in the three row so our main case to the MAX node +5 node. so it is change around to move at this +5 node, however we are select the move upward and backward position for that kind of looks good based on depending the backed-up main values of each of +5 node. some case the best node is value +5 (= max(+1,+5, 3)). So that minimax value base on the root node +5, and the move around select on this 3-ply in three in row. in other case our values to the next lower level -5 of each discs in the three row so our main case to the MIN node -5 node. so it is change around to move at this -5 node, however we are select the move backward and upward position for that kind of looks good based on depending the backed-up main values of each of -5 node. some case the best node is value -5 (= min(-1, -5, 3)). So that minimax value base on the root node -5, and the move around select on this 3-ply in three in row.

Simply define & justifying a heuristics. it like a rule of that people those depend on use for judgments and make own decisions. those are rules is generally depend on base of work as well as under most important & circumstances work, but some time they can follow to proper systematic deviations in that kind of logic also probability. those are mental clear the shortcuts of base on judgments that are generally involve the focus on a difficult complex problem and some time ignoring others people.

Related Questions

Navigate

• Browse (All)
• Browse S
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.