I have an optimization algorithm which is somehow similar to Bi-level optimization problem. Let's assume there is an algorithm (Alg1) form the literature as the state of the art for the problem.
I designed my own algorithm to solve the problem, but it doesn't provide a solution superior to Alg1, however if i use the Alg1's early solution (its first iteration) as the initial point of my algorithm, then it would converge to a better optimal point.
So how should i consider this regarding the novelty and how to represent my algorithm in general?