Nt1310 Unit 3 Quiz

1.8) Just as breathing and blinking involve a series of complex steps through complicated processes, our visual system has become effective and developed through evolution over millions of years. While the vision system may not be exactly performing this kind of mathematics, it has evolved in such a way that its design allows it to mimic this complex process. The gap between our visual systems abilities and our cognitive capabilities also helps free the conscious part of the mind, which allows animals to pay more attention to threats and be more likely to survive.

1.9) Evolution favors organisms that are able to live the longest and able to reproduce the most. Animals that acted irrationally were more likely to be killed and less likely to

The rational agent would have to keep track of where it has cleaned so that it could minimize plan the shortest route to minimize costs.

2.2 c) A simple agent vacuum could just randomly move around since it always the geography is unknown and new dirt may come back. A more complex agent, would learn the geography, what squares had dirt and how many movements it has taken since it has visited each square so that it can try to predict the likelihood of dirt reappearing and try to maximize the probability of finding more dirt. Learning allows the agent to give itself a more accurate model of its surroundings than a random model, which allows it to maximize its performance score.

2.3 a) False, the example simple vacuum agent only has partial information about the state of the things being clean and dirty but it will always perfectly rational since it always does the “right thing” at each step.

2.3 b) True, if a two agents were playing a game of copying where the reflex agent had to copy the previous move the second agent made, the reflex agent would not be

Second, the description also leaves out information about the robot, such as how long the robots battery life is.

Third, the description also leaves out information about the robot’s memory capabilities. It is unknown if the robot is able to remember every path it has taken and create a map of the maze or if it must continuously make reflex decisions.

3.3 a) Goal: Have the two friends meet in the same city, (i,i), in shortest time possible.

State space is the city pairs (i,j) where the two friends are.

3.3 b) Option (iii) Is admissible since in the best situation, in which both friends head directly for each other, the friends will be going twice as far as the cost of the time of each step.

3.3 c) No there are not completely connected maps for which no solution exists, assuming that one friend can stay in the same city during any transition, since there maps are connected there is a guaranteed path from every city to every other city.

3.3 d) No, since an optimal solution would never involve doubling back to a city more than once since it does not contribute to the solution nor provide a more optimal