arXiv:2406.18612v1 Announce Sort: new
Summary: This paper investigates the issue of regression mannequin era. A mannequin is a superposition of primitive features. The mannequin construction is described by a weighted coloured graph. Every graph vertex corresponds to some primitive perform. An edge assigns a superposition of two features. The burden of an edge equals the likelihood of superposition. To generate an optimum mannequin one has to reconstruct its construction from its graph adjacency matrix. The proposed algorithm reconstructs the~minimal spanning tree from the~weighted coloured graph. This paper presents a novel resolution primarily based on the prize-collecting Steiner tree algorithm. This algorithm is in contrast with its options.
Supply hyperlink