PROBABILITY HIRE- ASISTANT [closed] - algorithm

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 2 years ago.
Improve this question
Here is an algorithm for the HIRE-ASSISTANT problem.
HIRE-ASSISTANT(n)
best <- 0
for i <- 1 to n do
if candidate[i] is better than candidate[best]
best <- i
hire candidate i
Now some observations:
1.Candidate 1 is always hired.
2.The best candidate,i.e., the one whose rank is n, is always hired.
3.If the best candidate is candidate 1, then that is the only candidate hired.
Now the problem is what is the probability of hiring twice?
My approach:
Now before nth rank candidate, I can interview any number of candidates as I want to but the order of their rank is fixed.Therefore for i candidates being interviewed before nth rank candidate=C(n-1,i)*(n-i-1)! total cases are possible.So varying i=1 from n-1 and summing and dividing by total possibilities that are n! I calculate the answer, but it does not match with the standard answer so I need help in finding what is wrong?

The probability to hire At least twice is (n-1)/n.
Assuming you have a random permutation of the candidates, you hire only one candidate if and only if the first candidate is also the best. The probability for it to happen is 1/n. Thus, the probability that it won't happen (and you will hire twice or more) is 1-1/n = (n-1)/n
To hire exactly twice:
Choose a location (not first) for the 'best': n-1 possibilities. Let the place be i
For each such place, choose i-1 candidates to be placed before the best: Choose(n-1,i-1)
The first from these i-1 candidates is the best out of them. Need to permute the rest: (i-2)!
In addition, still need to permute for (n-i-1) candidates after the 'best': (n-i-1)!.
This gives us the total number of 'valid' permutations that exactly two candidates are hired are:
f(n) = Sum[ Choose(n-1,i-1)*(i-2)!*(n-i-1)! | for i=2,...,n]
And the probability is simply P=f(n)/n!

Actually, in the 4th bullet we need to permute (n-i) candidates so the final answer reduces to P = 1/n*H_(n-1) where H_(n-1) is the (n-1)^th harmonic number.

Related

Fast hill climbing algorithm that can stabilize when near optimal [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 5 years ago.
Improve this question
I have a floating point number x from [1, 500] that generates a binary y of 1 at some probability p. And I'm trying to find the x that can generate the most 1 or has highest p. I'm assuming there's only one maximum.
Is there a algorithm that can converge fast to the x with highest p while making sure it doesn't jump around too much after it's achieved for e.x. within 0.1% of the optimal x? Specifically, it would be great if it stabilizes when near < 0.1% of optimal x.
I know we can do this with simulated annealing but I don't think I should hard code temperature because I need to use the same algorithm when x could be from [1, 3000] or the p distribution is different.
This paper provides an for smart hill-climbing algorithm. The idea is basically you take n samples as starting points. The algorithm is as follows (it is simplified into one dimensional for your problem):
Take n sample points in the search space. In the paper, he uses Linear Hypercube Sampling since the dimensions of the data in the paper is assumed to be large. In your case, since it is one-dimensional, you can just use random sapling as usual.
For each sample points, gather points from its "local neighborhood" and find a best fit quadratic curve. Find the new maximum candidate from the quadratic curve. If the objective function of the new maximum candidate is actually higher than the previous one, update the sample point to the new maximum candidate. Repeat this step with smaller "local neighborhood" size for each iteration.
Use the best point from the sample points
Restart: repeat step 2 and 3, and then compare the maximums. If there is no improvement, stop. If there is improvement, repeat again.

Possible Array combinations based on constraints [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 6 years ago.
Improve this question
How many unique arrays of m elements exist such that they contain numbers in the range [1,n] and there exists atleast one subsequence {1,2,3,4....n}?
Constraints: m > n
I thought of combinations approach. But there will be repetitions.
In my approach, I first lay out all the numbers from 1 to n.
For example, if m=n+1, answer is n^2. (n spots available, each number in range [1,n])
Now, I think there might be a DP relation for further calculation, but I am not being able to figure it out.
Here's an example for n=3 and m=5. The green squares are the subsequence. The subsequence consists of the first 1 in the array, the first 2 that's after the first 1, etc. Squares that aren't part of the subsequence can either take n values if they are after the end of the subsequence, or n-1 values otherwise.
So the answer to this example is 1*9 + 3*6 + 6*4 = 51, which is easily verified by brute force. The coefficients 1,3,6 appear to be related to Pascal's triangle. The rest is left to the reader.

Average case of Linear search [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 3 years ago.
Improve this question
I have an array of elements A1,A2,...,An.
The probability of a user searching for each element are P1,P2,...,Pn.
If the elements are rearranged, will the average case of the algorithm change?
Edit : I have posted the question, which appeared in my exam.
The expected number of comparisons is sum_{i=1...n}(i * p_i).
Re-ordering the elements in descending order reduces the expectation. That's intuitive since by looking at the most probable choices first will reduce, on average, the number of elements looked at before you find a particular choice.
As an example, suppose there's three items k1, k2, k3 with match probabilities 10%, 30% and 60%.
Then in the order k1, k2, k3, the expected number of comparisons is 1*0.1 + 2*0.3 + 3*0.6 = 2.5
With the most likely first: k3, k2, k1, the expected number of comparisons is 1*0.6 + 2*0.3 + 3*0.1 = 1.5
No, because it takes O(1) time to access an element in array and it does not depend on a position of this element in array. So arr[0] and arr[10000] should take the same amount of time.
If you will have something like a linked list or a binary tree, then it makes sense to put elements that are accessed with higher probability closer to the beginning.

What's the minimal column sums difference? [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 7 years ago.
Improve this question
Imagine you are given a matrix of positive integer numbers (maximum 25*15, value of number does not exceed 3000000). When you do column sums and pick the smallest and the largest one, the difference between them must be the smallest possible.
You can swap numbers in every row (permute rows), not in column, how many times you want.
How would you solve this task?
I'm not asking for your code but your ideas.
Thanks in advance
I would make an attempt to solve the problem using Simulated Annealing. Here is a sketch of the plan:
Let the distance to optimize the difference between the max and min column sums.
Set the goal to be 0 (i.e., try to reach as close as possible to a matrix with no difference between sums)
Initialize the problem by calculating the array of sums of all columns to their current value.
Let a neighbor of the current matrix be the matrix that results from swapping two entries in the same row of the matrix.
Represent neighbors by their row index and two swapping column indexes.
When accepting a neighbor, do not compute all sums again. Just adjust the array of sums in the columns that have been swapped and by the difference of the swap (which you can deduce from the swapped row index)
Step 6 is essential for the sake of performance (large matrices).
The bad news is that this problem without the limits is NP-hard, and exact dynamic programming at scale seems out of the question. I think that my first approach would be large-neighborhood local search: repeatedly choose a random submatrix (rows and columns) small enough to be amenable to brute force and choose the optimal permutations while leaving the rest of the matrix undisturbed.

Powers of a half that sum to one [closed]

Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 8 years ago.
Improve this question
Call every subunitary ratio with its denominator a power of 2 a perplex.
Number 1 can be written in many ways as a sum of perplexes.
Call every sum of perplexes a zeta.
Two zetas are distinct if and only if one of the zeta has as least one perplex that the other does not have. In the image shown above, the last two zetas are considered to be the same.
Find all the numbers of ways 1 can be written as a zeta with N perplexes. Because this number can be big, calculate it modulo 100003.
Please don't post the code, but rather the algorithm. Be as precise as you can.
This problem was given at a contest and the official solution, written in the Romanian language, has been uploaded at https://www.dropbox.com/s/ulvp9of5b3bfgm0/1112_descr_P2_fractii2.docx?dl=0 , as a docx file. (you can use google translate)
I do not understand what the author of the solution meant to say there.
Well, this reminds me of BFS algorithms(Breadth first search), where you radiate out from a single point to find multiple solutions w/ different permutations.
Here you can use recursion, and set the base case as when N perplexes have been reached in that 1 call stack of the recursive function.
So you can say:
function(int N <-- perplexes, ArrayList<Double> currentNumbers, double dividedNum)
if N == 0, then you're done - enter the currentNumbers array into a hashtable
clone the currentNumbers ArrayList as cloneNumbers
remove dividedNum from cloneNumbers and add 2 dividedNum/2
iterate through index of cloneNumbers
for every number x in cloneNumbers, call function(N--, cloneNumbers, x)
This is a rough, very inefficient but short way to do it. There's obviously a lot of ways you can prune the algorithm(reduce the amount of duplicates going into the hashtable, prevent cloning as much as possible, etc), but because this shows the absolute permutation of every number, and then enters that sequence into a hashtable, the hashtable will use its equals() comparison to see that the sequence already exists(such as your last 2 zetas), and reject the duplicate. That way, you'll be left with the answer you want.
The efficiency of the current algorithm: O(|E|^(N)), where |E| is the absolute number of numbers you can have inside of the array at the end of all insertions, and N is the number of insertions(or as you said, # of perplexes). Obviously this isn't the most optimal speed, but it does definitely work.
Hope this helps!

Resources