# What is the name of the optimization algorithm that uses random sampling?

I am generating random weight as per e.g. below. The I generate a set of 3 values say 100, 250, 300 and I multiple them with the weights below

Initial population.

1. 0.1 0.7 0.2

2. 0.4 0.5 0.1

3. 0.3 0.5 0.2

4. 0.6 0.1 0.3

5  0.3 0.3 0.4

6. 0.2 0.2 0.6


Multiplication of weight

1. 0.1*100 + 0.7*250 + 0.2*300 = 245

2. 0.4*100 + 0.5*250 + 0.1*300 = 195

3. 0.3*100 + 0.5*250 + 0.2*300 = 215

4. 0.6*100 + 0.1*250 + 0.3*300 = 175

5. 0.3*100 + 0.3*250 + 0.4*300 = 225

6. 0.2*100 + 0.2*250 + 0.6*300 = 250


So after the multiplication we selected the individual 6 as the best combination of weights. This method works fine for us but we are trying to link it to any available algorithm. Does this algorithm have a specific name? Any idea or help on this? Is this not a genetic algorithm?

• Hi biz14 and welcome to scicomp! It's a bit difficult to understand what problem you're working on and what your specific question is. Rewriting it with more explicit detail will help us to help you better. – Paul Jan 6 '14 at 20:24
• @Paul sorry for my question I have rewrite with clearer steps and problem. Thank you. – biz14 Jan 7 '14 at 4:12
• It is clear that your objective is to find the maximum dot product between "populations" and "weights" given a fixed vector of "weights". However, your constraints are not clear. Are you restricting yourself to find a maximum only among the randomly generated "populations"? Or are you really looking for a maximum among all feasible values of "population" triplets? – Paul Jan 7 '14 at 18:06
• My constraint is that the weight are calculated using a sum formula from 3 different input which I think pretty much straight forward. Yes my restriction is the finding best among the population of the triplets ? So what is your suggestion? – biz14 Jan 8 '14 at 13:58
• My answer below remains the same. – Paul Jan 8 '14 at 15:09