Solution of \(n\)-person games by a proximal method of prediction type (Q5951123)

The author considers general \(n\)-person games where the set of strategies is described by nonlinear constraints. This type of game problem is then transformed into an appropriate saddlepoint problem. The author points out, that a straightforward solution of this saddlepoint problem by differential equations fails. The main part of the article is the convergence proof for a refinement of this method which uses feedback control (treated as prediction) and gives a solution to the original \(n\)-person game.