3 ms·

I get that they are generalizations; that is why I'm asking for specific instances. Say you were trying to develop a general algorithm to solve NP-complete pro

I get that they are generalizations; that is why I'm asking for specific instances. Say you were trying to develop a general algorithm to solve NP-complete problems in P-time. You want this algorithm so that you can solve actual, real-world problems. Now imagine that you only have enough computing power to solve 1 problem per week; you'd likely want to prioritize the problems you solve based on the 'value' derived from solving them. I'm asking, what is the problem with the highest value? Where could you have the most impact? Optimizing global shipping routes? Power-generation scheduling in Arizona? Encryption breaking?

ankurdhama 12y ago
The problem with the highest value will be very subjective as everyone from different fields will say that their problem is of highest value. I would suggest that you can pick a problem from a domain that you are most familiar with to understand the problem and it's impact. Wait... Are you asking about a start-up idea ;)
- jsprogrammer 12y ago
  That's a good point. I'm looking for highest value by the metric of something like 'weighted sum of subjective value of all problems'. Not really about a start-up idea. I've been talking to a person who claims to have such an algorithm, but is looking for an actual instance of a real-world problem. So far, no one I've asked has been able to describe such a problem.