Solving nonograms using Neural Networks

Each header indicates the number of cells that must be marked in a row inside the board to construct a block. If there is more than one number in the same row or column header, at least one empty cell must exist between them. Puzzles of an arbitrary size can be defined as rectangular or square. The cells of a nonogram are defined by two states: filled (| |) and empty (| x |). Figure 1: Examples of different nonogram states: unsolved, partially solved, and solved. The black cells are considered as filled, whereas those with a cross are empty. Figure 1 depicts the three stages of nonogram resolution: unsolved, partially solved, and solved. Note that this type of problem falls into the category of NP completeness [1, 2, 3]; thus, a solution cannot be obtained in polynomial time. Moreover, certain nonograms do not have a single solution, and all solutions that are compatible with the constraints defined by their headers are valid. An example of the situation is illustrated in Figure 2.