Transfer Learning for Underrepresented Music Generation

Combinational creativity, also sometimes combinatorial network models for music generation have arisen, trained creativity, is a type of creative problem solving on massive datasets and requiring significant computation in which two conceptual spaces are combined to represent (Civit et al. 2022). While these approaches have proven a third or new conceptual space (Boden 2009). While different successful at replicating genres of music like those in their musical genres may vary in terms of their local features training sets, due to the nature of large-scale neural network (e.g., melodies), they are all still music. As such, we models we expect this may not prove true for dissimilar genres.