Toward A Neuro-inspired Creative Decoder
This work addresses the challenge of enhancing creativity in AI systems for applications in art and design, though it appears incremental by building on existing deep generative methods.
The authors tackled the problem of generating novel and meaningful artifacts by proposing a neuro-inspired creative decoder that modulates neuronal activation patterns in a deep generative framework, resulting in improved creativity as measured by novelty metrics and human evaluation across multiple image datasets.
Creativity, a process that generates novel and meaningful ideas, involves increased association between task-positive (control) and task-negative (default) networks in the human brain. Inspired by this seminal finding, in this study we propose a creative decoder within a deep generative framework, which involves direct modulation of the neuronal activation pattern after sampling from the learned latent space. The proposed approach is fully unsupervised and can be used off-the-shelf. Several novelty metrics and human evaluation were used to evaluate the creative capacity of the deep decoder. Our experiments on different image datasets (MNIST, FMNIST, MNIST+FMNIST, WikiArt and CelebA) reveal that atypical co-activation of highly activated and weakly activated neurons in a deep decoder promotes generation of novel and meaningful artifacts.