general questions to know Flashcards
what is depth-first search
it will follow the path from the starting node to the end and then again from the start to the end until all the nodes are visited uses a stack structure
what is breadth-first search
proceeds level by level visiting all the nodes from one level and uses a queue structure
what is admissible heuristics
admissible: never overestimate the goal
explain the hill climbing algorithm
given a current state , then look at all the successor nodes then pick the one with the best cost
what is the downside of hill climbing
only improves on the given solution and gets stuck on the local minima
explain the simulated annealing
the idea is that it looks like hill climbing :
so where : T=0 => acts like hill climbing
when T= infinity , behaves randomly
Hence when t drops there are less random movement
hence when thetemperature decreases slowly enough the simulated anealing sercah finds the global optimum propbility
what is a 0 sum game
for a given two players in a gum if one of them wins then the other loses
what is the difference between adversarial and non-adversarial game
search is adversary where the solution is a heuristic
while games are adversaries and a solution is a strategy
explain the MinMax game setup
there are two players Max and Min, one wins and the other receives penalty
if it is taken as a search algorithm then we have:
initial state: board configuration
successor function: list specifying the legal moves
terminal state: has the game ended
Utility function: Gives numerical value of terminal states
in this type of game, we want to maximize our moves and minimise our opponent’s move
what is the issue with e minmax algorithm an how to improve
the minmax goes all over the tree , one solution is to user the (alpha,beta pruning)
How to check whether a system os markov
if the actionas of the whole system depends solely on the the current action , if a system has the markov property then it is considered as a MArkov decision process
how to selec action in RL
epsilon greedy or through using the Boltzmann equation
how alpha and beta affect the ant colony algorithm and rho
alpha plays an important role in the generation of the pheromone while beta plays an important role on selecting high quality solution
rho plays role in the decrease of the pheromone where where small value lead to exploitive behaviour while large values lead to expletive behaviours
How to eavluate the serach strategy
completness, optimatlity , complexity
what is the difference between informed and uninformed strategy
usually uninformed one uses only the information on problem formulation while an informed one relies on heuristics
what are some optimal serach algo
Best first serach and Depth first seach only if the cost is 1 per step
what is best firs tsearch
it is a serach algorithm that combines the functionalities of BFS and DFS
