sequential convex programming

Sequential convex programming (SCP) is an iterative optimization method used to solve non-convex optimization problems. The method involves dividing a non-convex problem into a sequence of convex subproblems, which are then solved iteratively.

At each iteration, the non-convex problem is approximated by a convex subproblem, which is then solved using standard convex optimization techniques. The solution of the subproblem is used to update the solution of the non-convex problem, and the process is repeated until convergence is achieved.

SCP is particularly useful in solving optimization problems with non-convex constraints or objective functions, where traditional optimization methods may fail to find a global optimum. The method has been widely used in machine learning, control theory, and other fields where optimization is a key component.

One potential drawback of SCP is that it may converge to a local optimum rather than a global optimum. However, this can be mitigated by starting the algorithm from multiple initial points and choosing the best solution among them.