0
$\begingroup$

I am working on a learning to rank problem. I have queries and documents related to every query which I have to rank. I used lightgbm ranker to fit the model. Some of features are very important and if they are changed the fitted model predicts a better score for that document and thus a better rank.

Lets say, for a single query id, I have a group of documents d1....d5 each having features f1...fn. I change the features f1,f2,f3 for document d4 and it pushes the rank of d4 from 4 to 3.

  1. What should be the values of these features f1,f2,f3 that can push the rank for document d4 from 4 to 1 subject to the condition that f1+f2+f3=45 is always satisfied ? (i.e. the sum of new modifications for f1,f2,f3 should be always equal to 45 )

As far as I understand this problem is simply a constrained optimization problem. In such a problem we have an objective function and constraints. In my case I do not have a mathematical equation that relates rank to my features r=f(f1....fn) but I do have a learned model which I can use as an objective function.

Follow up question

All this made me think of another question. In the case of OLS, which is a constrained optimization problem, the constraints present in this kind of optimization are always on the coefficients.

  1. If I would like to have constraints on the predictors does it really make sense ? How does it relate to my problem ?

I think the objective function of a regression problem and the objective function of constrained optimization problem i mentioned above are different. I would really appreciate some thoughts on this, thanks !

$\endgroup$
0
$\begingroup$

To be honest I didn't understand the problem very well, but it looks like genetic learning could be a good direction to study.

Essentially a genetic algorithm works by iteratively trying many combinations of values for some features, evaluating "something" based on the features, selecting the best combinations and starting again with a combination of the best features so far. The goal is to eventually converge to an optimal combination of values.

$\endgroup$
1
  • $\begingroup$ thanks for the reply ! I edited my question $\endgroup$ – Siddhant Tandon Dec 10 '20 at 7:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.